Intelligent Systems

Computational Intelligence

teach

 

Vorlesung (infCi-01a)

2 SWS
Zeit und Ort: Mi 18:00 - 19:30, LMS2, Raum Ü1
vom 29.10.2019 bis zum 2.2.2020
 

Übung (ÜinfCI-01a)

2 SWS
Zeit und Ort: Mi 18:00 - 19:30,LMS 2, Raum Ü1
vom 29.10.2019 bis zum 2.2.2020

 

Kurzfassung

Der Begriff "Computational Intelligence" (CI) beschreibt ein Teilgebiet der künstlichen Intelligenz. Im Wesentlichen fasst er drei biologisch motivierte Fachgebiete der Informationsverarbeitung zusammen: Basierend auf Algorithmen der Fuzzylogik und künstlichen neuronalen Netzen sowie auf den Evolutionären Algorithmen wird die Beherrschung komplexer Systeme angestrebt und mit weiteren, typischerweise biologisch-inspirierten, Verfahren kombiniert. Urspünglich in den 1990er-Jahren vom Institute of Electrical and Electronics Engineers (IEEE) geprägt, wird der Begriff heutzutage oft synonym mit dem Soft-Computing verwendet. Alle beinhalteten Teilgebiete haben gemeinsam, dass sie Mechanismen natürlicher (d. h. insbesondere biologische, physische oder soziale) Problemlösungsstrategien für mathematische oder Ingenieur-technische Fragestellungen nutzbar machen. Dabei wird nicht auf eine direkte Übertragung oder "technische Kopie" abgezielt, sondern auf ein Verständnis und eine Imitation der Basismechanismen. Die so entwickelten Verfahren stehen im Gegensatz zu exakt-mathematischen Verfahren - man folgt eher frei dem Motto: "Erlaubt ist, was funktioniert".

Lernziele

Ziel der Veranstaltung ist die Vermittlung eines ersten Überblicks über das Gebiet der Computational Intelligence in Theorie und Praxis. Darauf aufbauend sollen Studierende ein Grundverständnis und entsprechende Herangehensweisen vermittelt werden, sodass insbesondere folgende Ziele erreicht werden:

  • Die Studierenden haben ein Grundverständnis der Komplexität von Technischen Systemen und wissen wie diese beherrscht werden kann.
  • Sie wissen wie scheinbar komplexe Zusammenhänge mit Mechanismen der Computational Intelligence einfach beschreiben werden können.
  • Techniken aus dem Feld Computational Intelligence / Soft Computing sowie deren Vor- und Nachteile im Vergleich sind bekannt.

 

Workload

60 Std. Vorlesung, 30 Std. Präsenzübung, 150 Std. Selbststudium

   

Intelligent Systems

Lecture (infInS-01a)

4 SWS, ECTS studies, ECTS credits: 8

language of instruction: English
Time and place: Wednesday 10:15 - 11:45 a.m., Ludwig-Meyn-Str. 2, room Ü3, Thursday 12:15 - 13:45 p.m., WSP2, room 214
from October 20, 2019 to February 2, 2020
Exam / examination: February 19, 2020, 9:00 a.m. - 11:30 a.m., room CAP3 - lecture hall 1
oral review on March 25th, 2020 appointment by arrangement with the secretariat

Exercise (ÜinfInS-01a)

2 SWS
Unterrichtssprache Englisch
Time and place: Monday 18:00 - 19:30 p.m., LMS2, room Ü1
from November 11, 2019 to January  27, 2020

 

Summary

An "Intelligent system" is a computing system capable of operating under difficult conditions (e.g. time-varying environments, emergent situations or disturbances) by autonomously adapting its behaviour to changing conditions and learning autonomously. The main goal of engineering intelligent systems is to counter the challenges of complexity by means of integrating desired characteristics such as robustness, flexibility or resilience into technical systems. This is combined with a continuous improvement of the system behaviour. The improvement process is achieved by different approaches of machine learning, e.g. from the fields of reinforcing, active or semi-supervised learning. Besides these learning-related aspects, the design and organisation of large-scale intelligent systems consisting of a potentially large group of autonomous subsystems requires techniques for self-organisation as well as mechanisms for trust relations and fairness.

The lecture gives an introduction to the design and realisation of intelligent systems. It is based on the insights of research initiatives such as "Organic Computing" and "Autonomic Computing".

 

Learning goals

The overall goal of the course is to derive a basic understanding of the motivation, the general concept, and engineering methods of intelligent systems. Based on this, students will learn about machine learning techniques capable of gathering and describing the environmental and internal conditions of an intelligent systems as well as for improving the behaviour autonomously at runtime.

Particular goals are:

  • Students understand the motivation and the need for intelligent systems that act autonomously without (or with only limited) user intervention or guidance.
  • Students can define the terms "Intelligent System", "Organic Computing", and "Autonomic Computing"
  • Students are able to design intelligent systems by assessing and selecting a suitable basic model.
  • Students can implement selected methods for clustering and classifying situations based on data gathered by sensors.
  • Students can compare algorithms for learning from feedback and implement the most promising variant.
  • Students are capable of quantifying system aspects of large-scale organisations of autonomous intelligent systems with respect to characteristics such as robustness, emergence, self-organisation, autonomy, or adaptivity.

 

Workload

60 h lecture, 30 h exercise, 150 h self-study

ECTS

8

   

Master Project Intelligent Systems

Exercise  (infMPInS-01a)

 

4 SWS, ECTS-Credits: 10
suitable for ERASMUS / exchange students, language of instruction: English
Time and place: by arrangement
 

Turtle Bot

Scope

This project is offered as a master project – a modified variant for bachelor students is possible (please ask)!

 

Group

This project is offered as a group task; a team of 3 to 6 students is required (deliverables are adjusted based on a number of participants).

 

Language

The project is available in English and German, depending on the participants.

Motivation: Robots are a perfect testbed for intelligent systems technology – and autonomous learning techniques in particular. Their mobility, their heterogeneity, and their different application areas define a variety of use cases where learning capabilities are required to initially solve and afterwards improve certain tasks.

 

Task

What is your task as a group?

  • Set up the basic robot operating system for the robots, assemble robots (i.e., attach sensors, access sensor information, process sensor information).
     
  1.  We have the turtle bot 3 platform available at our group, which is used for this module. This includes the “TurtleBot3 Waffle Pi” and the “TurtleBot3 Burger” platforms with additional sensor/actuator equipment.
  2. There are a very popular framework and API available that need to be adapted for the scope of this project (framework and API come with comprehensive documentation, tutorials, and community).
  3. The API / framework are available in Python and C++, experimental libraries are also available in JAVA. The programming language is subject to negotiations.
     
  • Design and implement a framework for autonomous robot behaviour:
     
  1. Works on heterogeneous robots (i.e., waffle and burger bots with varying sensor and actuator equipment)
  2. Is modular in a way that novel intelligent behaviours can easily be added (and switched on/off)
  3. Follows the design concepts for intelligent systems and provides basic functionality for robot operation tasks
  4. Provides continuous logging of system states, environmental conditions, and actions taken for analysis purposes
  5. Consists of a variable set of autonomous robots and a “base station” with which the robots communicate and share their experiences

 

  • Use case: classification and novelty detection of environmental conditions
  1. Realise (hardware) scenarios for novel behaviour
    a) Here: environmental behaviour is modelled as different flooring/surface conditions
    b) Set up courses with changing conditions (carpet, stones, tiles, etc)
     
  2. Classification and novelty detection techniques
    a) Implement classification of floor conditions based on sensor information representations
    b) Implement techniques for novelty detection (abnormal behaviour) that generate an alarm in case of inappropriate classification (i.e., unknown floor)
    c) Apply techniques online and classify floor conditions with different robots
    d) Evaluate the success of the classification

     

Fundamentals of the module

  • Besides the actual development of the software, the module aims at practical experiences in the following aspects:
    - Specification and definition of “product” using standard software engineering tools
    - (Self-)Organisation as a team, management of the process, deadline supervision
    - Entire software development process until delivery to the customer
    - Documentation of the product
  • Your supervisors are continuously monitoring the process in the role of a customer, i.e. you are expected to regularly demonstrate the progress.
     

Registration

Interested students may contact the group (i.e., Prof. Tomforde at st@informatik.uni-kiel.de) for more details.

There will be an initial appointment begin of the summer term, where the project is set up and students are registered for the module.

Selbst-organisierte Systeme

Bachelorseminar (infBSemIS-01a)

Seminar, 2 SWS, ECTS-Studium, ECTS-Credits: 4
für ERASMUS-/Austauschstudierende geeignet, Unterrichtssprache Englisch
Zeit und Ort: n.V.
vom 6.4.2020 bis zum 5.7.2020
 

Kurzfassung

Das Seminar widmet sich dem Thema "Selbst-organisierte Systeme". Darunter werden im Kontext dieses Seminars Mengen aus autonomen und technischen Einheiten verstanden, die zusammenarbeiten, um ein gemeinsames Ziel zu erreichen. Dabei treten unterschiedliche Fragestellungen auf, die beherrscht werden müssen.

Das Seminar im Wintersemester 2019/20 widmet sich dabei der konsistenten, schnellen und zuverlässigen Speicherung sowie des damit einhergehenden Auffindens von Inhalten innerhalb solcher selbst-organisierten Systeme. Dies wird anhand von Algorithmen des Peer-to-Peer Computing untersucht.

Das Seminar im SS 20 widmet sich dabei der Kooperation von autonomen intelligenten Systemen sowie den damit einhergehenden Interaktionsmustern innerhalb solcher selbst-organisierten Systeme. Dies wird anhand von Beispielen des Autonomic und Organic Computing untersucht.

 

Lernziele

Die Studierenden lernen, sich komplexe Sachverhalte anhand der kompakten Darstellung, wie sie in wissenschaftlichen Veröffentlichungen (d. h. Paper, Artikel oder Fachbüchern) üblich ist, eigenständig anzueignen und in verständlicher Form aufzubereiten.

 

Workload

30 Std. Mitarbeit im Seminar, 90 Std. Selbststudium