Lecture with exercise for Bachelor AI / ISE CEProgrammieren mit C

Lecturer: Prof. Dr. Gregor Schiele (Lecture)
Lukas Einhaus (Exercise)
Language: German
Turnus: Summer Term
Time: Wednesday, 10:15-12:00 Uhr (Lecture)
Tuesday, 12:00-14:00 Uhr (Exercise)
Location: LC 137
Start: 13.04.2022

This Bachelor lecture teaches the basics of programming in the C programming language. Despite its long history, C is still one of the most widespread and important programming languages today, especially for system-oriented programming. It is easy to learn but difficult to master, as it contains only a few keywords and concepts, but these can be used to emulate many modern programming techniques.

In detail, it covers: General concepts of programming languages, variables and types in C, operators and expressions, control structures and functions, the preprocessor, pointers, static and dynamic memory management, error handling, bit manipulation, modules and abstract data types, unit testing and test-driven development in C.

Please note that this is not a lecture on the basics of programming. The lecture builds on the courses "Grundlegende Programmiertechniken" and "Fortgeschrittene Programmiertechniken", i.e. basic programming knowledge (e.g. variables and types, loops, subroutines and recursion) and basic knowledge of data structures and algorithms are assumed.

Lecture with exercise for Master AI / ISE CE / ISE ESEInternet of Things: Protocols and System Software

Lecturer: Prof. Dr. Gregor Schiele (Lecture)
Chao Qian (Exercise)
Language: English
Turnus: Summer Term
Time:

Tuesday, 16:00-18:00h

Wednesday, 14:00-16:00h

Location:

Tuesday: LE 105

Wednesday: LC 137

Start: 05.04.2022

This masters lecture will introduce the research area of the "Internet of Things" (IoT), where billions of embedded systems (actuator, sensors) make continously data in real time to about the real world available to the Internet. Topics, that will be covered are:

Hardware plattforms (e.g. SBCs, Sensor nodes), communication protocols (e.g. IEEE 802.15.4, 6LoWPAN, CoAP, MQTT), data modelling (e.g. linked data, RDF, SSN), data management and access (e.g. SPARQL, continous queries with CQELS, "Big Data"), system software and software plattforms (e.g. Eclipse Ponte, Xively, BASE, PCOM), future work ("programmable world"). An exercise will complement the theoretical knowledge with practical task of programming IoT-Systems, e.g. with Arduinos or Raspberry Pis.

Student ProjectStudent Project Embedded Systems - Bio-signal processing for exoskeletons SS22

Supervisor: Prof. Dr. Gregor Schiele
Christopher Ringhofer
Chao Qian
Language: German/English
Turnus: Summer Term
Time: 05.04.2022 - 10:00 - 11:00 Uhr (Kickoff)
Location: BC 013
Beginn: TBD

Medical exoskeletons are an important topic e.g. for medical rehabilitation of stroke patients. One challenge is to identify what kind of movement a user wants to do, e.g. lifting an arm. This can be supported using biosignal analysis such as EEG (i.e., neuron activity in the brain) or EMG (i.e., muscle activity). In this project, we will explore how embedded machine learning can be used to implement better exoskeletons through such biosignal analysis. 

 
We offer this project in cooperation with the medical technology department of Prof. Elsa Kircher, who is working with medical exoskeletons and will provide access to an existing system that will be used as our starting point and comparison system. In addition, the project will be done in parallel to a project for medical technology, who will provide input on biosignal analysis. 
 
Project members will
  • explore the state of the art for biosignal analysis for exoskeleton control,
  • design and train machine learning (ML) algorithms for biosignal analysis (EEG, EMG),
  • use an existing toolchain to implement embedded accelerators for such ML algorithms,
  • evaluate the developed system with real experiments.
The project is planned as the first part of a one year (two semesters) project.
 
This project is open for: BAI, MAI, and MCPS
 
Prerequisite: 
- Experience with C/C++ programming
 
Advantageous:
- Prior knowledge of time series analysis and / or digital signal processing
- Experience with designing / training machine learning solutions
- Experience with embedded systems development
- Python or VHDL programming experience
 

Student ProjectPraxisprojekt Eingebettete Systeme - Embedded AI Model Store

Supervisor: Prof. Dr. Gregor Schiele
Lukas Einhaus
Language: German/English
Turnus: Summer Term
Time: 04.04.2022 - 14:00 - 16:00 Uhr (Kickoff)
Location: BC 013
Beginn: TBD

Embedded AI is found in many fields today, often allowing efficient processing of data close to the data source. By their nature, however, embedded systems are often hardly physically accessible. Since they are also subject to changing requirements and environmental influences, application administrators and developers need a way to make changes to the AI even after the system has been deployed. In this project, the first steps will be taken to update heterogeneous embedded AI in-situ or even to change it completely. To this end, the students will develop a solution that allows an embedded system to access an AI model store. They will

  • research how similar systems are realised today
  • design/train an example model for an embedded MCU and integrate it into an example application
  • set up a model store
  • store models and parameters in the model store
  • develop interfaces that allow application administrators to access entities in the model store.

Prerequisite for participation:

  • Knowledge of embedded C or Python

Helpful knowledge:

  • Java
  • MQTT
  • Pytorch

The kick-off meeting will take place on 04.04.2022 at 14:00 in room BC013. Attendance of the kick-off is mandatory for participation in the project.

Bachelor and Master SeminarSeminar „Reconfigurable Hardware for the Internet of Things“

Supervisor: Prof. Dr. Gregor Schiele
Christopher Ringhofer
Lukas Einhaus
Language: English/German
Turnus: Summer Term
Time: Do, 10:00-12:00h
Location: BC 303
Beginn: Do, 07.04.2022, 10:00 (Kickoff)

In order to be able to cope with the upcoming challenges of applications in the Internet of Things, further hardware classes must be included that can solve certain problem classes of computations more efficiently. Reconfigurable types of hardware such as field-programmable gate arrays (FPGA) represent an alternative to the classically used microcontrollers. The topic of this seminar is the development of applications for reconfigurable hardware with a focus on FPGAs. Not only will the theoretical basics of FPGAs and the development language VHDL be highlighted, but practical work will also be done with these devices in the seminar and they will be programmed in VHDL in order to directly and practically deepen the theoretically learned concepts. During the seminar, an extended abstract is to be written and an application written in VHDL is to be developed.


The kick-off meeting will take place on 07.04.2022 at 10:00 am in room BC303. Attendance of the kick-off meeting is mandatory for participation in the seminar.

Practical TrainingCyberphysical Systems Labor-Praktikum

Teacher:

Prof. Dr. Gregor Schiele / Chao Qian
Prof. Dr. Torben Weis / Peter Zdankin
Prof. Dr. Josef Pauli / Martin Moder

Language: German
Turnus: Summer Term
Time: TBD
Location: TBD
Start: Mi, 06.04.2022, 10:00 (Kickoff) BC013

This practical training is exclusive for students of the course of study M.Sc. "Cyber Physical Systems". It is offered in collaboration with the groups of Prof. Pauli, Prof. Weis and Prof. Schiele.

Kickoff: Mi, 06.04.2022 10:00 BC013