PC1: Programming Autonomy - Practical -

Speakers:

Hana Boukricha and Nhung Nguyen

Disciplines:

Computer Science, Artificial Intelligence & Computer Graphics

Duration:  

4 sessions

Organization:
Number of participants is limited to 20. Enrollment is on a
first-come-first-serve basis at the IK. The practical excercises
should be solved in groups of at least 2 persons.
As technical requirements participants are requested to bring
own Laptops with either operating systems Windows, Linux, or
Mac OS X installed and at least 100 MB of free disk space.
(For additional technical information please visit the
website listed in Reference 2) This course addresses participants with basic programming skills.

Course content:
Multiagent systems seem to be a natural metaphor for understanding
and building a wide range of what we might crudely call
artificial social systems. The field of multiagent system deals with two following problems (see p. 3, Reference 1):

  • Building agents capable of independent autonomous actions, in order to successfully carry out the tasks delegated to them.
  • Building agents that are capable of interacting with other agents in order to successfully carry out the tasks delegated to them, particularly when the other agent can not be assumed to share the same interests or goals.

Using a 3D Computer Graphics framework, agents based on simple reactive
systems will be implemented. The agents are situated in a virtual world
in which they interact with the environment as well as with each other. We start with implementing single agent behaviors and extend them to competetive and cooperative multiagent behaviors. As main scenario the "Mars Explorer Experiment" from Luc Steels as described in Reference 1, is chosen.
At the end a tournament between the multiagent teams will take place.

  • Session 1: Building Braitenberg Vehicles in breve (see Reference 2).
  • Session 2 and 3: Implementing the Mars Explorer Experiment (see chapter 5, Reference 1) in breve.
  • Session 4: Tournament of the multiagent teams.

Objectives:

  1. Acquiring practical experiences in the field of multiagent systems and
    autonomous agents by implementing theoretical multiagent concepts.
  2. Developing and enhancing agent architectures to cope with environments
    growing in their complexity.


Literature:
 

  1. Michael Wooldridge, Introduction to MultiAgent Systems, John Wiley and Sons Ltd, February 2002
  2. Software Framework and additional material:
    http://www.spiderland.org

Vita:
Besides working towards their Ph.D. theses, Hana and Nhung enjoy showing undergraduate students how to provide agents with simple behaviors so that they (the agents) can get along in graphically simulated but yet challenging 3D worlds.

Hana Boukricha:

"http://www.techfak.uni-bielefeld.de/~hboukric/"
Hana Boukricha studied "Applied Computer Science in the Natural Sciences" at Bielefeld University.  Currently she is a Ph.D. student in the Artificial Intelligence Group headed by Prof. Dr. Ipke Wachsmuth. Furthermore, she is working as a research assistant in the Collaborative Research Center 673 "Alignment in Communication".  In her doctoral thesis she is investigating how to endow a virtual human with the ability to "empathize" with its interaction partner by allowing him affective partner modeling using facial mimicry and role-taking.

Nhung Nguyen:

"http://www.techfak.uni-bielefeld.de/~nnguyen/"
Nhung Nguyen studied Computer Science with a second major in physics at Bielefeld University. Since 2007 she works as a research assistant in the Artificial Intelligence Group of Prof. Dr. Ipke Wachsmuth. In her Ph.D. thesis she investigates how artificial, embodied (virtual) agents use different bodily sensor modalities to represent their own body and their surrounding space.

 

Last update: 26.01.2011, Webadmin