Human-Agent Cooperation with ACT-R Based Intent Recognition in a Dynamic Environment
- Bachelorthesis -
Context and Motivation
Artificial intelligence and autonomous agents are becoming an increasingly important part of our everyday and social environment. As humans and artificial agents will increasingly need to cooperate, the ability to understand a partner’s intentions and provide appropriate assistance may become essential for successful cooperation.
Goal of the Thesis
In this thesis, an intelligent cooperative agent will be created for the Cooperative Cuisine environment. The agent will use intent recognition to infer the intentions of its human cooperation partner from their observable actions and adapt its own behavior accordingly. The thesis will investigate how intent recognition influences the success of human-agent cooperation and whether it can contribute to greater trust in the agent.
Evaluation
The evaluation will consider objective measures such as task completion, completion time, and errors, as well as subjective measures such as trust and perceived reliability. It
will also examine how mistakes made by the agent affect the human partner’s trust and
willingness to rely on the agent.
Anforderungen/Kenntnisse:
Kognitive Modellierung mit ACT-R, Anbindung von ACT-R an Python, Durchführung einer Evaluierung
Bearbeitung:
Nevena Taneva
Betreuung:
Prof. Dr. rer.nat. Nele Rußwinkel
Institut für Informationssysteme
Ratzeburger Allee 160 ( Gebäude 64 - 2. OG)
23562 Lübeck
Telefon: 0451 / 3101 5700
