Natural language-based human-robot control

With the increase in the number of robots arriving at our homes, it is important to find an easy and efficient way to communicate with them. Natural language is the most natural form of communication between humans, so its use by users to communicate with robots, with no knowledge of how they work,...

Full description

Bibliographic Details
Main Author: Bernardo, João Miguel Nunes (author)
Format: masterThesis
Language:eng
Published: 2020
Subjects:
Online Access:http://hdl.handle.net/10071/20093
Country:Portugal
Oai:oai:repositorio.iscte-iul.pt:10071/20093
Description
Summary:With the increase in the number of robots arriving at our homes, it is important to find an easy and efficient way to communicate with them. Natural language is the most natural form of communication between humans, so its use by users to communicate with robots, with no knowledge of how they work, is an added value to control them. We present a solution that allows us to interact with a robot in a natural way, so it is easier for human-robot cooperation to happen. This solution includes a method of transforming the information in natural language into controls that a robot can perform. We implement this solution in the specific case of indoor navigation. Finally, we evaluate the performance and interactivity of our approach, by having users, without previous knowledge of robotic control, controlling a robot in a simulated environment.