Daphne: an Intelligent Cognitive Assistant for the design of Earth observing satellites
Tutor / Supervisor
Selva Valero, Daniel
Student
Prat i Sala, Arnau
Document type
Master thesis
Date
2017
rights
Open Access
Publisher
Universitat Politècnica de Catalunya
UPCommons
Abstract
This thesis introduces Daphne, an Intelligent Cognitive Assistant (ICA) designed to help engineers architect Earth observing satellite systems. The purpose of this thesis is threefold: 1) develop the interfaces through which the user interacts with the system, 2) develop the basic intelligence of the system and 3) perform a preliminary experiment to evaluate the effectiveness of the system in improving design outcomes. Three different interfaces have been developed which include different modes of interaction, visual interfaces (including virtual reality) as well as a physical embodiment for non-verbal interaction. An artificial intelligence has been developed to help users explore the design space by providing criticism and suggestions about the design the user is currently working on. Finally a preliminary experiment has been performed to measure the effectiveness of the system. Ultimately, the goal of Daphne is to enable mixed-initiative human-computer design of complex systems, in which humans and Intelligent Cognitive Assistants engage in balanced collaborations: not dominated by the human or the computer but rather engaging in a meaningful dialogue. This project is part of a research grant awarded by the United States NSF (National Science Foundation).
Entitat col·laboradora
Cornell University
Participating teacher
- Selva Valero, Daniel