Algorithms for AI Robotics
Lectures/Exercises: Pavel Petrovic, I24a, http://dai.fmph.uniba.sk/~petrovic/ , petrovicfmph.uniba.sk.
TOTO JE ARCHÍVNA STRÁNKA Z PREDCHÁDZAJÚCICH ROKOV, SLÚŽI LEN NA ORIENTÁCIU
Aktuálna stránka: https://dai.fmph.uniba.sk/w/Course:AIRobotika/sk
The course will introduce the students to selected algorithms for intelligent robotic systems
and provide first-hand experience with experimental robotics in simulation and on real robots.
Grading
Project: 50 pts (working implementation of a method, algorithm, selected article or chapter)
Exam: 50 pts (30 pts for test covering the course material, 20 pts for essay - compiled topic sk/en, min. 5 pages)
A: 90+
B: 80-90
C: 70-80
D: 60-70
E: 50-60 (but at least pts from project + 10)
Exercises
In the lab: anytime in the assigned slots (Wed, Fri: 8-12), you need to book your presence in the calendar (total min. 10 hours work),
alternately: at home/another lab, but report what you did when.
Lab reservation calendar
Lab rules
Exams
Will be announced in the second part of semester.
How to prepare? You can bring one A4 sheet containing your notes, and no other materials. Please try to read through the material (see link somewhere else on this page), the questions will refer to it, when needed, you will get the necessary article, or part of it. The exam will include computational task on prob. robotics (adjusting belief based on specification of the sensor), and updating value or Q-function in reinforcement learning, and some other tasks.
Example exams: example1, example2.
Lectures
- Introduction, overview of basic concepts I, slides, notes
- Introduction and overview II, slides
- Review of LEGO, navigation, SBOT, slides
- EA + RL = LCS, slides
- LCS, NEAT, slides
- Fly algorithm, CMAES, MOEA, slides
- ER, BBR, CBR, slides
- MDP, slides
- POMDP, slides
- SIFT & SURF, see the articles, Hough transform - see tutorial opencv
- Probabilistic Robotics, Bayesian Robot Programming, slides1, slides2
- Quaternions
The material covered may include:
- Perception and sensor systems,
- Software robotic architectures
- Representation and inference in space
- Navigation and localisation
- Probabilistic approaches
- Logic approaches to robot programming
- Simulation of robotic systems
- Robotics and artificial life (applications of evolutionary algorithms and neural networks for robotics)
- Applications and other topics
The materials are at a separate page: (uzivatel airob heslo je meno robotickej sutaze, ktora bola koncom aprila odzadu) material.
Our department is for the third year building a common laboratory with the
Institute of Robotics and Industrial Informatics of Slovak Technical University.
The aim of this course is to cover interesting algorithms that can be useful in the field
of Robotics (but with some modifications also in other applications).
The students will be introduced to the practical work with the robots in the laboratory,
and in the second part of the course, the students will work on a specific small-size
project in groups. The lab meetings are arranged individually.
Projects
Recommended readings
- Murphy: Introduction to AI Robotics, MIT Press, 2000
- Thrun, Burgard, Fox: Probabilistic Robotics, MIT Press, 2005.
- Dudek, Jenkin: Computational Principles of Mobile Robotics, Cambridge Univ. Press, 2000
- Kortenkamp, Bonasso, Murphy, Artificial Intelligence and Mobile Robots, MIT Press, 1998
- Nehmzow, Mobile Robotics: A Practical Introduction, Springer, 2000
- Bruce, Green, Georgeson: Visual Perception, 4th ed, Psychology Press, 2003.
- Nolfi, Floreano: Evolutionary Robotics, MIT Press, 2000
- Kalaš, Tridsať rokov svetovej robotiky, Vydavatelstvo STU, 2006