Control of robot swarms through natural language dialogue: A case study on monitoring fires

There are numerous environmental and non-environmental disasters happening throughout the world, representing a big danger to common people, community helpers, to the fauna and flora. Developing a program capable of controlling swarms of robots, using natural language processing (NLP) and further on...

Full description

Bibliographic Details
Main Author: Mendonça, Nuno Miguel Amorim (author)
Format: masterThesis
Language:eng
Published: 2019
Subjects:
Online Access:http://hdl.handle.net/10071/20181
Country:Portugal
Oai:oai:repositorio.iscte-iul.pt:10071/20181
Description
Summary:There are numerous environmental and non-environmental disasters happening throughout the world, representing a big danger to common people, community helpers, to the fauna and flora. Developing a program capable of controlling swarms of robots, using natural language processing (NLP) and further on, a speech to text system, will enable a more mobile solution, with no need for keyboard and mouse or a mobile device for operating with the robots. Using a welldeveloped NLP system will allow the program to understand natural languagebased interactions, making this system able to be used in different contexts. In firefighting, the use of robots, more specifically drones, enables new ways to obtain reliable information that before was based on guesses or knowledge from someone who had long-time experience on field. Using a swarm of robots to monitor fire enables innumerous advantages, from the creation of a dynamic fire map, climate information inside the fire, to finding lost firefighters on field through the generated map. This work uses firefighting as a case-study, but other situations can be considered, like searching someone in the sea or searching for toxins in an open environmental area.