Hasso-Plattner-Institut25 Jahre HPI
Hasso-Plattner-Institut25 Jahre HPI

Machine Learning-based Control of Dynamical Systems (Sommersemester 2021)

Dozent: Prof. Dr. Holger Giese (Systemanalyse und Modellierung) , Christian Medeiros Adriano (Systemanalyse und Modellierung) , Sona Ghahremani (Systemanalyse und Modellierung) , He Xu (Systemanalyse und Modellierung)

Allgemeine Information

  • Semesterwochenstunden: 4
  • ECTS: 6
  • Benotet: Ja
  • Einschreibefrist: 18.03.2021 - 09.04.2021
  • Lehrform: Projekt / Seminar
  • Belegungsart: Wahlpflichtmodul
  • Lehrsprache: Englisch

Studiengänge, Modulgruppen & Module

IT-Systems Engineering MA
  • IT-Systems Engineering
    • HPI-ITSE-K Konstruktion
  • IT-Systems Engineering
    • HPI-ITSE-M Maintenance
  • SAMT: Software Architecture & Modeling Technology
    • HPI-SAMT-K Konzepte und Methoden
  • SAMT: Software Architecture & Modeling Technology
    • HPI-SAMT-T Techniken und Werkzeuge
  • SAMT: Software Architecture & Modeling Technology
    • HPI-SAMT-S Spezialisierung
  • OSIS: Operating Systems & Information Systems Technology
    • HPI-OSIS-K Konzepte und Methoden
  • OSIS: Operating Systems & Information Systems Technology
    • HPI-OSIS-T Techniken und Werkzeuge
  • OSIS: Operating Systems & Information Systems Technology
    • HPI-OSIS-S Spezialisierung
Data Engineering MA



As technology behind AI systems are becoming more complex, society is faced with the problem of understanding and explaining the outcomes of these systems, for instance, deep learning-based systems which were shown to be brittle and easy to trick. 

The technical solutions have been to build models for explaining AI systems. However, explainability does not scale in mission-critical systems, in which split-second decisions have to be made.  Take for instance when one has to intervene (control) to avoid a traffic collision or the exponential spread of a virus. These are decisions that have to be made in real-time by air-traffic control or autonomous vehicles systems  and socio-technical systems platforms to monitor disease spread. An alternative approach is to design robustness and resilience into mission-critical systems. 

While systems built traditionally rely on instantaneously explainability and perfect human understanding, we will study how to build feedback control systems that provide robustness under the conditions of current mission-critical domains. 

We will study methods that combine classical and machine learning-based control. These methods support the discussion of various design concerns, for instance, how to extend classical methods beyond navigation to manipulation domains? Since the world is not static, how should we build robust systems over changing environments? How can two-way consistency be utilized beyond the scope of graphical-based planning methods? 


  • Learn the fundamentals and the methods to implement feedback control systems using machine learning techniques.
  • Learn how to apply tools to integrate feedback control in mission-critical systems that must be steered and stabilized over extended horizons of time.


[1] Brunton, Steven L., and J. Nathan Kutz, (2019), Data-driven science and engineering: Machine learning, dynamical systems, and control. Cambridge University Press.

[2] Kutz, J. Nathan, et al., (2016), Dynamic Mode Decomposition: Data-Driven Modeling of Complex Systems. Vol. 149. SIAM.

[3] Duriez, Thomas, Steven L. Brunton, and Bernd R. Noack, (2017), Machine learning control, taming nonlinear dynamics and turbulence. Vol. 116. Heidelberg: Springer.

Lern- und Lehrformen

The course is a project seminar, which has an introductory phase comprising initial short lectures. After that, the students will work in groups on jointly identified experiments applying specific solutions to given problems and finally prepare a presentation and write a report about their findings concerning the experiments.

There will be an introductory phase to present basic concepts for the theme including the necessary foundations.


We will grade the group's experiments (60%), reports (30%), and presentations (10%). Participation in the project seminar during meetings and other groups' presentations in the form of questions and feedback will also be required.


After the introductory phase with few initial short lectures, we will identify the group topics and then there will be regular individual feedback rounds of the groups with their supervisors. In addition, there will also be regular meetings during the semester for the whole project seminar to discuss the progress of all groups and open questions in general.

The introductory meeting will be held online on Tuesday 20.04 at 11:00 am.

Regular meetings: Tuesday at 11:00, Wednesday at 15:15

Please email christian.adriano(at)hpi.de to obtain the Zoom credentials. If you are interested in the project seminar but cannot attend the introduction meeting, please email us. We will find an individual solution for you.

Announcement regarding the coronavirus regulations: 

Because of the Coronavirus situation and corresponding restriction outbreak, we will organize all meetings as online meetings by default. This especially applies to the first lecture. If all participants agree to and the current restrictions as well as seminar room availability allow it, further meetings may also take place at HPI.