Smart cities leverage advancements in big data and artificial intelligence to deliver a multitude of services and information to urban people. Among these services, predicting on-street parking availability is an important application with the potential to enhance parking efficiency, alleviate city congestion, and minimize pollution. Existing methods for forecasting parking occupancy rates mostly rely on recurrent neural networks (RNNs) to capture temporal dimension information from parking time series data. However, these methods typically overlook the crucial spatial dependency among parking areas, resulting in suboptimal prediction accuracy. Furthermore, the computationally intensive nature of RNN-based methods leads to slow prediction speeds. To address these limitations, we propose Multi-view Spatial-temporal Graph Convolutional Networks (MV-STGCN) to predict parking occupancy rates. By integrating spatial and temporal features, MV-STGCN is able to capture complex spatial-temporal correlations and improve prediction accuracy while optimizing prediction speed. The proposed MV-STGCN incorporates a multi-view contrastive Graph Convolution module (mvc-GConv), which employs a multi-view contrast method to extract features from topology and feature spaces with commonalities and differences in a multi-view way. Experimental results based on real-world datasets demonstrate that MV-STGCN outperforms baselines in predicting long-term parking occupancy rates while achieving superior prediction speed.

Spatial database systems and Geographic Information Systems (GIS) are mainly able to support geographical applications that deal with crisp spatial objects , that is, objects whose extent, shape, and boundary are precisely determined. But geoscientists have pointed out for a long time that there is also a need to represent fuzzy spatial objects that reveal an intrinsically vague or blurred nature and structure and feature indeterminate boundaries and/or interiors. A spatial object is fuzzy if locations exist that cannot be assigned completely to the object or to its complement. In this article, we propose an abstract, formal, and conceptual type system called Fuzzy Spatial Algebra ( FUSA ) that provides a collection of fuzzy spatial data types for fuzzy points , fuzzy lines , and fuzzy regions in the two-dimensional Euclidean space. We introduce a set of expressive spatial operations such as fuzzy union , fuzzy intersection , and fuzzy difference to perform geometric computations on fuzzy spatial objects. As a specialty, users may exert influence on how spatial fuzziness is interpreted and handled in these operations. Our formal framework is based on fuzzy set theory and fuzzy topology. FUSA is designed to serve as a specification of its implementation in a spatial database and GIS context. We show the applicability of FUSA and its possible embedding into the query languages of extensible database systems by employing a running example.

Trajectory prediction aims to estimate an entity’s future path using its current position and historical movement data, benefiting fields like autonomous navigation, robotics, and human movement analytics. Deep learning approaches have become key in this area, utilizing large-scale trajectory datasets to model movement patterns, but face challenges in managing complex spatial dependencies and adapting to dynamic environments. To address these challenges, we introduce TrajLearn , a novel model for trajectory prediction that leverages generative modeling of higher-order mobility flows based on hexagonal spatial representation. TrajLearn predicts the next k steps by integrating a customized beam search for exploring multiple potential paths while maintaining spatial continuity. We conducted a rigorous evaluation of TrajLearn , benchmarking it against leading state-of-the-art approaches and meaningful baselines. The results indicate that TrajLearn achieves significant performance gains, with improvements of up to ~40% across multiple real-world trajectory datasets. In addition, we evaluated different prediction horizons (i.e., various values of k ), conducted resolution sensitivity analysis, and performed ablation studies to assess the impact of key model components. Furthermore, we developed a novel algorithm to generate mixed-resolution maps by hierarchically subdividing hexagonal regions into finer segments within a specified observation area. This approach supports selective detailing , applying finer resolution to areas of interest or high activity (e.g., urban centers) while using coarser resolution for less significant regions (e.g., rural or uninhabited areas), effectively reducing data storage requirements and computational overhead. We promote reproducibility and adaptability by offering complete code, data, and detailed documentation with flexible configuration options for various applications.

The accurate and prompt mapping of flood-affected regions is important for effective disaster management, including damage assessment and relief efforts. While high-resolution optical imagery from satellites during disasters presents an opportunity for automated flood inundation mapping, existing segmentation models face challenges due to noises such as cloud cover and tree canopies. Thanks to the digital elevation model (DEM) data readily available from sources such as United States Geological Survey (USGS), terrain guidance was utilized by recent graphical models such as hidden Markov trees (HMTs) to improve segmentation quality. Unfortunately, these methods either can only handle a small area where water levels at different locations are assumed to be consistent or require restricted assumptions such as there is only one river channel. This article presents an algorithm for flood extent mapping on large-scale Earth imagery, applicable to a large geographic area with multiple river channels. Since water level can vary a lot from upstream to downstream, we propose to detect river pixels to partition the remaining pixels into localized zones, each with a unique water level. In each zone, water at all locations flows to the same river entry point. Pixels in each zone are organized by an HMT to capture water flow directions guided by elevations. Moreover, a novel regularization scheme is designed to enforce inter-zone consistency by penalizing pixel-pairs of adjacent zones that violate terrain guidance. Efficient parallelization is made possible by coloring the zone adjacency graph to identify zones and zone-pairs that have no dependency and hence can be processed in parallel, and incremental one-pass terrain-guided scanning is conducted wherever applicable to reuse computations. Experiments demonstrate that our solution is more accurate than existing solutions and can efficiently and accurately map out flooding pixels in a giant area of size 24,805 × 40,129. Despite the imbalanced workloads caused by a few large zonal HMTs dominating the serial computing time, our parallelization approach is effective and manages to achieve up to 14.3× speedup on a machine with Intel Xeon Gold 6126 CPU @ 2.60 GHz (24 cores, 48 threads) using 32 threads.

As the volume and complexity of spatiotemporal data continue to expand rapidly across various domains such as urban planning, environmental monitoring, and logistics, the demand for comprehensive data management systems becomes increasingly urgent. Handling such data entails intricate topological and analytical operations, emphasizing the necessity for robust and adaptable solutions capable of addressing diverse user queries. This article introduces Distributed MobilityDB, 1 an open source system engineered to manage big spatiotemporal trajectory datasets within SQL environments. Distributed MobilityDB offers capabilities for scalable spatiotemporal data management, facilitating efficient distributed query processing while seamlessly integrating with existing MobilityDB SQL operations. Key contributions highlighted in the article encompass an adaptive spatiotemporal SQL query engine. This engine channels user SQL queries through various planning strategies for optimizing the distributed query plan, then distributing the query execution across cluster nodes transparently to the user. Various spatiotemporal query types are supported for distribution, including range selections, and joins proximity. Distributed MobilityDB is implemented as an add-on extension to PostgreSQL, which facilitates installing it on a readily running server. The article further presents extensive experiments conducted on both cloud and on-premise environments using both real and synthetic datasets, including the Automatic Identification System for ship trajectories and BerlinMOD for simulated person trips.

Polygon overlay operations are used for various purposes, such as GIS searches and queries, VLSI, and basic geometric operations of intersection, union, and difference. There have been recent research articles presenting algorithms using the GPU to perform line segment intersection for geometric operations. We present two parallel algorithms implemented on the GPU that focus on the active list portion of the traditional serial plane sweep algorithm. The first algorithm uses a single block of threads to simulate the active list data structure in hardware; this algorithm is slow due to GPU thread block size limitations and synchronization points but demonstrates favorable time complexity. The second algorithm uses dynamic parallelism to remove synchronization and scales to utilize available GPU hardware (single GPU). We perform experiments on both synthetic and real-world datasets. The presented results show improvement in execution time with respect to recent algorithms and low memory usage compared to recent algorithms. We achieve speedups of up to 38.8 over the serial sweep line algorithm on real-world data.

Application domains such as environmental health science, climate science, and geosciences—where the relationship between humans and the environment is studied—are constantly evolving and require innovative approaches in geospatial data analysis. Recent technological advancements have led to the proliferation of high-granularity geospatial data, enabling such domains but posing major challenges in managing vast datasets that have high spatiotemporal similarities. We introduce the Hierarchical Grid Partitioning (HierGP) framework to address this issue. Unlike conventional discrete global grid systems, HierGP dynamically adapts to the data’s inherent characteristics. At the core of our framework is the Map Point Reduction algorithm, designed to aggregate and then collapse data points based on user-defined similarity criteria. This effectively reduces data volume while preserving essential information. The reduction process is particularly effective in handling environmental data from extensive geographical regions. We structure the data into a multilevel hierarchy from which a reduced representative dataset can be extracted. We compare the performance of HierGP against several state-of-the-art geospatial indexing algorithms and demonstrate that HierGP outperforms the existing approaches in terms of runtime, memory footprint, and scalability. We illustrate the benefits of the HierGP approach using two representative applications: analysis of over 289 million location samples from a registry of participants and efficient extraction of environmental data from large polygons. While the application demonstration in this work has focused on environmental health, the methodology of the HierGP framework can be extended to explore diverse geospatial analytics domains.

Similarity search is the problem of finding in a collection of objects those that are similar to a given query object. It is a fundamental problem in modern applications and the objects considered may be as diverse as locations in space, text documents, images, X (formerly known as Twitter) messages, or trajectories of moving objects. In this article, we are motivated by the latter application. Trajectories are recorded movements of mobile objects such as vehicles, animals, public transportation, or parts of the human body. We propose a novel distance function called DistanceAvg to capture the similarity of such movements. To be practical, it is necessary to provide indexing for this distance measure. Fortunately we do not need to start from scratch. A generic and unifying approach is metric space, which organizes the set of objects solely by a distance (similarity) function with certain natural properties. Our function DistanceAvg is a metric. Although metric indexes have been studied for decades and many such structures are available, they do not offer the best performance with trajectories. In this article, we propose a new design, which outperforms the best existing indexes for kNN queries and is equally good for range queries. It is especially suitable for expensive distance functions as they occur in trajectory similarity search. In many applications, kNN queries are more practical than range queries as it may be difficult to determine an appropriate search radius. Our index provides exact result sets for the given distance function.

High-quality, typically administrative, geospatial data should adhere to established measurement and representation practices and be protected from malicious attacks. However, this kind of geospatial data may only be infrequently updated due to its often prolonged production process compared to a data source of volunteered geographic information such as OpenStreetMap. Existing approaches typically try to quality-assure geospatial data by comparing it to another reference dataset of perceived higher quality - often another administrative dataset facing a similar update cycle. In contrast, this article tries to determine whether actual changes present in volunteered geographic information data such as OpenStreetMap, which also need to be applied in an administrative dataset (i.e., consists of actual changes in the real world), can be identified automatically. To that end, we present QPredict, a machine learning approach observing changes in volunteered geospatial data such as OpenStreetMap to predict issues with a target (administrative) dataset. The algorithm is trained by exploiting geospatial object characteristics, intrinsic and extrinsic quality metrics and their respective changes over time. We evaluate the effectiveness of our approach on two datasets representing two mid-size cities in Germany and discuss our findings in terms of their applicability in use cases.