Collaborative agents and perspective taking via visual situated analyzing
- Bachelorarbeit -
Description:
Day by day, our interaction with intelligent agents and robots is becoming increasingly essential not only in daily life but also in professional contexts, hence intelligent agents should be capable of taking the human perspective into account to provide effective support.
The thesis explores a solution in the form of an intelligent agent that assists users in complex visual search tasks, such as navigating hidden object scenes. To fulfill that goal, the agent must understand the human behavior based on real-time glances, which are captured with the eye tracking glasses. Based on that, it will correct the human and navigate him through verbal guidance to reach the needed object.
The prototype is implemented and tested in a controlled environment. In addition, the experience of the participants will be considered to evaluate the effectiveness of the agent in supporting and reducing search time.
The following question will be examined:
Can a robot infer a human's target based on glances and provide tailored assistance accordingly?
Anforderungen/Kenntnisse:
Python-Programmierung, Integration von Eye Tracking Glasses, Durchführung einer Evaluierung
Bearbeitung:
Nihad Beddeiwi
Betreuung:
Prof. Dr. rer.nat. Nele Rußwinkel
Institut für Informationssysteme
Ratzeburger Allee 160 ( Gebäude 64 - 2. OG)
23562 Lübeck
Telefon: 0451 / 3101 5700
