CORE Module “Intelligent Systems”

This module teaches you about core technologies and algorithms which endow a man-made system with intelligence. You will learn how machines can process sensor data, including visual data, to perceive and represent their surroundings. Once an environment representation is available, an intelligent machine, such as a robot, can act on and change its environment after deliberate planning. Utilizing its accumulated experience, the machine can learn and adapt its behavior in the future. This module covers all of these aspects and thus gives you an in-depth understanding of machine learning, computer vision, as well as robotics.


Computer Vision

Computer Vision algorithms are used in a variety of real-world applications, such as surveillance and object tracking, 3D model building (photogrammetry), and object recognition. Apart from their visual appeal, these algorithms also represent elegant applications of linear algebra and optimization techniques. Topics covered in this course include a recapitulation of relevant linear algebra, introduction to face-recognition, camera calibration, stitched panoramas, edge and blob visual features, structure from motion, color-spaces, segmentation, basic 3D point-cloud processing, and an introduction to object-recognition. It is assumed that the student can program in C and Matlab.

  • Semester: Fall (3rd semester)
  • ECTS: 5
  • Instructor: Prof. Dr. Kaustubh Pathak

Machine Learning

Machine learning (ML) is about algorithms which are fed with (large quantities of) real-world data, and which return a compressed “model” of the data. An example is the “world model” of a robot: the input data are sensor data streams, from which the robot learns a model of its environment — needed, for instance, for navigation. Another example is a spoken language model: the input data are speech recordings, from which ML methods build a model of spoken English — useful, for instance, in automated speech recognition systems. There exist many formalisms in which such models can be cast, and an equally large diversity of learning algorithms. However, there is a relatively small number of fundamental challenges which are common to all of these formalisms and algorithms. The lecture introduces such fundamental concepts and illustrates them with a choice of elementary model formalisms (linear classifiers and regressors, radial basis function networks, clustering, neural networks, hidden Markov models). Furthermore, the lecture also provides a refresher of required mathematical material from probability theory and linear algebra.

  • Semester: Spring (4th semester)
  • ECTS: 5
  • Instructor: Prof. Dr. Herbert Jaeger

Robotics

The course gives an introduction to robotics with a particular focus on (intelligent) mobile robots. The lecture covers the according core methods and technologies to enable autonomous or semi-autonomous operations of mobile platforms. Examples of related topics include

  • actuators and their components, i.e., electrical motors, gears, feedback sensors
  • locomotion, especially different drive units, their physical implementation and their kinematics
  • robot control architectures
  • localization sensors and methods
  • range sensing and processing
  • map representations
  • core principles of Simultaneous Localization and Mapping (SLAM)
  • sensor data registration
  • place recognition
  • obstacle avoidance and introduction to path-planning
  • Semester: Spring (4th semester)
  • ECTS: 5
  • Instructor: Prof. Dr. Andreas Birk