M. Sc. Daniel Flögel
- Research Assoiate
- Group:
FZI
- Room: FZI
- Phone: +49 721 9654-175
- Fax: FZI
- floegel ∂ fzi de
FZI Forschungszentrum Informatik
Embedded Systems and Sensors Engineering (ESS)
Haid-und-Neu Str. 10-14
76131 Karlsruhe
Curriculum Vitae
Studies of electrical engineering and information technology at the Karlsruhe Institute of Technology (KIT) with majoring in control technology. Active membership in the Formula Student team KA-RaceIng with responsibility for data acquisition in the season 2016 and team leadership of electronics department in the season 2017. Practical experience at Mercedes-AMG in the development of high-performance high voltage batteries and subsequent bachelor thesis on electro-thermal modeling and performance optimization of high performance cooled high voltage batteries for hybrid vehicles (2019).
Research assistant at the Forschungszentrum für Informatik (FZI) on function development for visualizations of driver assistance functions. Master thesis at the University of Waterloo in Canada with the topic " Cooperative State Estimation for Autonomous Mobile Robots" (2022).
Since June 2022 research associate in the department Control in Information Technology (CIT) in the division of Embedded Systems and Sensors Engineering (ESS) at FZI.
Research
Human-machine interaction in the context of trajectory planning in high-density urban environments.
While highly autonomous driving on highways and city streets is a heavily researched area, new and previously unexplored challenges arise in urban environments. When highly autonomous machines, e.g., vehicles or mobile robots, move through high-density urban environments such as a pedestrian zone, sidewalks, or malls, complex intersection and avoidance scenarios occur between multiple dynamic objects and one machine. The focus of this research is on scenarios between multiple individually moving humans and one machine.
In such group scenarios, the autonomous machine must cooperate with the humans in a socially acceptable manner and make a joint collaborative interactive decision under mutual interaction. The decision-making process is complicated, among other things, by the fact that the participants move in a highly unstructured environment, without fixed lanes or generally applicable rules such as exist in road traffic, and there is no direct communication between humans and machines. In addition, there is uncertain information about the intention of the participants as well as new and so far not considered scenarios from the individual movements of the human.
The goal of this research is to design a trajectory planning in dynamic and unstructured environments with a focus on decision making and interaction between humans and the machine. The requirements for the planning are a safe, both physically and psychologically, and comfortable trajectory for all involved while guaranteeing the achievement of the goal.
Title | Type | Supervisor |
---|---|---|
Multi-Objective Reinforcement Learning für die Bewegungsplanung von autonomen Robotern in Menschenmengen | Master Thesis |