DD2438 Artificial Intelligence and Multi Agent Systems 15.0 credits

Artificiell intelligens och multiagentsystem

In this course, we will study problems related to

  • AI for computer games (such as Call of Duty, FIFA14, Starcraft)
  • AI for multi robot search and rescue
  • AI for robot soccer
  • AI for domestic service robots

This includes topics such as

  • Cooperative path planning
  • Cooperative task assignment
  • Formation keeping
  • Motion coordination 

In computer games, as well as in future robot systems,groups of cooperative agents, so-called multi agent systems, bring new possibilities in terms of capabilities, efficiency and flexibility.

In this course you will design and implement a number of solutions to multi agent problems. You will be using simulation environments such as

  • Unity 3D (https://unity3d.com)
  • Unreal engine (https://www.unrealengine.com/)

That enable you to understand, test and develop you designs, in a polished environment, but also export them to stand alone games (if you want to).

The course will be given in English.

  • Education cycle

    Second cycle
  • Main field of study

    Computer Science and Engineering
    Information Technology
    Information and Communication Technology
  • Grading scale

    P, F

Course offerings

Spring 19 agent19 for programme students

Spring 18 agent18 for programme students FULL

Intended learning outcomes

After completing the course the students should:

  • be able to use a number of important tools and technologies used in artificial intelligence and multi agent systems
  • be able to develop intelligent multi-agent systems
  • be able to assess the value of, and to a suitable extent utilize, existing solutions as a part of a programming project
  • be able to plan and lead the work in a larger project
  • be able to present their work and results, both orally and in writing
  • be able to write a basic scientific paper in English.

Course main content

The students will in project form design and implement a multi-agent team performing a task. The actual course content can vary based on which solutions the students choose to use.
The following areas will to a smaller or greater extent, dependent on the students' choices, be treated in the course: 

  • Cooperative path planning
  • Cooperative task assignment
  • Formation keeping
  • Motion coordination 

The course will also train the ability to manage, plan and participate in larger projects,  assess existing solutions and their possible use, and work with existing code.

Eligibility

KTH-students:
2D1345/DD1345, Introduction to Computer Science and 2D1240/DN1240, Numerical Methods, Basic Course II or 2D1241/DN1241 Numerical Methods, Basic Course III or equivalent.
2D1363/DD1363, Software Engineering or equivalent is recommended.

Single course students:
90 university credits including 45 university credits in Mathematics or Information Technology. English B, or equivalent.

Recommended prerequisites

DD2380 Artificial Intelligence (or a similar course)

Literature

To be announced at least 4 weeks before course start at course web page.

Examination

  • INL1 - Hand-in Assignment, 3.0, grading scale: P, F
  • PRO1 - Software Engineering Project, 4.0, grading scale: P, F
  • PRO2 - Software Engineering Project, 4.0, grading scale: P, F
  • PRO3 - Software Engineering Project, 4.0, grading scale: P, F

In this course all the regulations of the code of honor at the School of Computer science and Communication apply, see: http://www.kth.se/csc/student/hederskodex/1.17237?l=en_UK.

Offered by

CSC/Robotics, Perception and Learning

Contact

Petter Ögren, petter@kth.se, telefon: 790 6646

Examiner

Petter Ögren <petter@kth.se>

Version

Course syllabus valid from: Spring 2014.
Examination information valid from: Spring 2014.