Spatio: An Edge-Oriented Real-Time Spatial Data Engine for Mobility Systems

Read full text Summarise with VIRAT
Authors: Petro Kvartsianyi
Cited by: 0
References: 17

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Autonomous and connected mobility systems increasingly rely on onboard computation to operate safely and efficiently in dynamic, real-world environments. These systems continuously generate and process spatio-temporal data under strict constraints on latency, energy consumption, and computational resources. Many existing geospatial data management platforms provide rich spatial querying and analytical capabilities. Still, they are primarily designed for centralised or cloud-oriented deployment, often assuming stable network connectivity and abundant computational resources. Such assumptions limit their applicability in embedded and edge-based settings, where intermittent connectivity and constrained hardware are the norm. In this work, we present a spatio-temporal data engine for edge applications. The proposed engine is small, self-sufficient, designed for the ingestion of low-latency spatial data, allows self-contained agents to consider their real-time spatial environment, and is independent of the cloud infrastructure. It supports real-time positional updates, memory-based spatial indexing, and spatial queries for decision-making. The proposed system is implemented in Rust and provides memory safety, high, consistent performance, and low resource overhead, making it ready for deployment on embedded/edge-class hardware. The architecture is designed to be straightforward to integrate, exposing a compact, low-level API that can be embedded directly within higher-level control and perception pipelines. The engine is evaluated using synthetic mobility traces, measuring query latency, ingestion throughput, and memory utilisation under conditions representative of edge deployments. These experiments demonstrate that maintaining only the current spatial state of entities enables predictable memory usage and consistently low query latency under sustained update rates. In contrast to trajectory-oriented spatial databases, which prioritise historical completeness, the proposed engine adopts a state-based design that explicitly trades long-term storage for bounded resource consumption and deterministic performance.

Fields of Study & Subjects

Data Management and Algorithms Advanced Database Systems and Queries Geographic Information Systems Studies

Publication details

Journal / source
Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT)
Publication date
27 February 2026
Language
EN
Document type
Thesis