DSME Colloquium: Kai Dresia on fuel efficient rocket control with deep reinforcement learning
Thursday, 23rd of March 2023
On Thursday, 23rd of March 2023, at 14:00, we welcome Kai Dresia, Ph.D. student with the German Aerospace Center (DLR), at the DSME Colloquium where he will give a talk on "Fuel Efficient Rocket Engine Control with Deep Reinforcement Learning ", followed by an Q&A session. Please find the abstract of the talk and the biography of the speaker below.
This DSME Colloquium will be in a hybrid format, with the talk taking place at the DSME seminar room and streamed via Zoom. In order to receive the link to Zoom room, please subscribe to the DSME Colloquium mailing list firstname.lastname@example.org (by sending an email with subject "subscribe" to email@example.com). We will also announce future talks of the DSME Colloquium on this list.
You are welcome to attend the DSME Colloquium in person. If you do, please contact our office at firstname.lastname@example.org to register and receive directions.
Fuel Efficient Rocket Engine Control with Deep Reinforcement Learning
The talk will explore the intricate challenges involved in controlling rocket engines of reusable launch vehicles. These engines play a vital role in meeting the surging demand for launching new satellites in Earth's orbit. To overcome the challenges of reusability, machine learning algorithms such as reinforcement learning can be utilized. However, achieving the required robustness while training these algorithms in simulation presents significant difficulties. The presentation will showcase the outcomes of a real-world demonstration of a reinforcement learning controller at a rocket engine test bench at the DLR test site in Lampoldshausen. This demonstration will highlight the potential of machine learning algorithms for real-time control of rocket engines. Furthermore, it will emphasize the difficulties involved in implementing such systems in practical applications, such as robustness and safety aspects.
Kai Dresia received his master’s degree in aerospace engineering from RWTH Aachen University, Germany. He is a Ph.D. student for intelligent rocket engine control systems with the German Aerospace Center (DLR). The goal of his Ph.D. research is to develop an intelligent rocket engine controller for future liquid propellant rocket engines.