TU Delft
print this page print this page     
2010/2011 Electrical Engineering, Mathematics and Computer Science Master Computer Science
Seminar Human-Agent/Robot Teamwork
Responsible Instructor
Name E-mail
Dr. M.B. van Riemsdijk    M.B.vanRiemsdijk@tudelft.nl
Contact Hours / Week x/x/x/x
Education Period
Start Education
Exam Period
Course Language
Expected prior knowledge
IN4010 (Artificial Intelligence Techniques) is recommended, otherwise at least knowledge of Java
Course Contents
In this course, we study intelligent systems that cooperate with humans to achieve a common objective. These systems can consist of intelligent software agents, but may also be embodied in the form of robots. The theme of this course is thus human-agent/robot teamwork (HART). More specifically, the focus is on three main areas within HART:

- organizational modeling languages: languages for modeling an organization of agents and/or humans by specifying the roles that can be played, the norms that should be followed, etc.,
- (socio-cognitive) robotics: the study of (social) robots that are expected to become part of our daily lives, such as maids in our homes, robots that assist in challenging tasks such as surveillance or crisis rescue operations, caretaking assistants for the elderly, etc.,
- explainable AI (XAI): the study of how to explain to human users or teammates why certain decisions were taken by agents or robots and what line of reasoning was followed to come to this decision.

The course is a combination of a seminar and a practical assignment. In the first half of the seminar, main scientific papers on the topics described above will be discussed and analyzed. In the second half, students will present a paper of their choice on one of the above topics, and perform a literature study on the topic.

The practical assignment involves modeling an agent organization, programming a team of agents or simulated robots that can cooperate with humans, or programming an explanation facility. The purpose of the practical assignment is to get hands-on experience with the implementation of intelligent systems that cooperate with humans, and to compare this experience with the theory as presented in the papers. At the end of the course, a report has to be handed in in which the result of the practical assignment is described, and related to a literature study on the respective topic.
Study Goals
After succesful completion of the course, students can:
- analyze a scientific paper
- present a scientific paper
- explain key notions of HART frameworks
- analyze similarities and differences between HART frameworks
- develope (parts of) an intelligent system that cooperates with humans
- relate the developed intelligent system to the HART theory
- evaluate to what extent the theory supports the development of systems for HART
Education Method
Lectures, seminar, practical assignment
Computer Use
Students are expected to use their own computers for the practical assignment.
Literature and Study Materials
Each year, a new selection of important papers will be made. See Blackboard for the current selection.
- quality of presentation
- quality of report (including description of the developed system)
- participation