Type: Bachelor-/ Master Thesis Studien-/ Diplomarbeit
Die nachfolgende Aufgabenstellung ist in englischer Sprache verfasst.
Die Bearbeitung der Themen kann, je nach Wunsch des Studenten,
wahlweise in Englisch oder Deutsch erfolgen.
Gaze Control for Gathering Attention in Human Approach Scenarios
In order to make an autonomous robot more interactive it should approach humans pro-actively and initiate interaction instead of waiting for someone to start a conversation. To approach persons in a way that they respond to an interaction attempt, various aspects have to be taken care of. Attention has to be drawn to the robot and the robot intention of interacting has to be clear and readable.
The goal of this work is to teach the IURO robot behaviors of nonverbal communication. When approaching a human, IURO should use his gaze
to convey his intention. Yet, simply staring at the person that is being
approached is not sufficient. Models of nonverbal communication are to
applied to let the approach appear natural and human like. IURO should
also be able to switch his gaze direction between the tasks. During an
approach the gaze could shift from the person to the ground for traversability
assessment and further to salient areas in the environment.
To address this control problem, an algorithm should be developed that fuses
available information from sensors and algorithms and controls the mechatronic head/eyes in accordance to social communication models.
Approach proposal:
- Literature search
- Interfacing of 3 DoF Neck and 2 DoF Eye Hardware
- Integration of models for nonverbal communication
- Performance evaluation and testing via user studies
Supported Languages:
- German
- English
Prerequisites:
- Good programming skills in C/C++
Helpful but not required:
- Acquainted with Linux
- Experience with ROS
- Your own Laptop with a running Ubuntu version
Supervisor:
Literature:
- J. Batista, P. Peixoto, H. Araujo
Real-time vergence and binocular gaze control
Intelligent Robots and Systems, 1997. IROS '97 - T. J. Olson, D. J. Coombs
Real-Time Vergence Control for Binocular Robots
Technical Report, Series/Report Number: UR CSD / TR348 - C. Pelachaud, M. Bilvi
Modelling Gaze Behavior for Conversational Agents
Intelligent Virtual Agents (IVA), 2003