Embedded Files
Using UAVs and Retrieval Augmented Generation for Situational Awareness in Rescue Operations
on April 18, 2025 in Book chapter | LINK HERE
Abstract
The increasing frequency and severity of natural disasters in Brazil, particularly in 2024, highlight the urgent need for innovative solutions in Search and Rescue (SAR) operations. This work presents an approach which integrates Retrieval-Augmented Generation (RAG) techniques with Unmanned Aerial Vehicles (UAVs) to enhance real-time data processing, usability, and operator decision-making. By incorporating advanced technologies such as FrameNet Brasil, Robot Operating System 2 (ROS2), and Large Language Models (LLMs), the system transforms UAV-captured data into actionable insights accessible through natural language interfaces. Testing demonstrates its ability to improve situational awareness, identify critical points of interest, and streamline mission execution. This modular and scalable approach lays the groundwork for future advancements in SAR technologies and their application in disaster-prone regions.
Authors
Matheus B. Jenevain, Mario A. R. Dantas, Laís R. Berno, and Milena F. Pinto
A Context-Aware Approach to Data Exchange in the Energy Sector
on October 15, 2024 in Book chapter | LINK HERE
Abstract
This paper presents a novel approach to enhance data exchange within the energy sector by developing a context-aware method utilizing Natural Language Processing (NLP) techniques and ontologies such as Open Energy Ontology (OEO). The proposed solution addresses the challenges arising from differing contexts and perceptions, particularly among actors like ENTSO-E and its member countries. By capturing, representing, and assigning meanings based on the specific realities of the energy sector, this work seeks to minimize miscommunications and complications that may arise during data exchange. The approach is built upon previous research exploring ontology development and context perception
Authors
Matheus B. Jenevain, Mario A. R. Dantas, Laís R. Berno, and Milena F. Pinto
Enabling Intelligent Data Exchange in the Brazilian Energy Sector: A Context-Aware Ontological Approach
on April 11, 2024 in Book chapter | LINK HERE
Abstract
The ever-expanding requirement for interoperability between systems, coupled with the growing nu mber of participants in the energy sector, creates a scenario where diverse actors, each with unique realities and specificities, must exchange data and knowledge. This exchange often occurs among significant differences in context. These disparities can lead to many difficulties and misunderstandings, as the information exchanged is susceptible to misinterpretation based on the sender’s and receiver’s contexts. Therefore, this work addresses this issue by proposing an extension to the Open Energy Ontology (OEO) [1] that focuses on context. It investigates how an actor’s understanding is shaped by their context, the methods for inferring this context, and strategies to enhance interoperability. The results obtained demonstrate the potential of the approach proposed by this work.
Authors
Matheus B. Jenevain, Milena F. Pinto, Mario A. R. Dantas, Regina M. M. B. Villela, Jose M. N. David & Victor S. A. Menezes
Page updated
Google Sites
Report abuse