SOS Team Members

Kai Cui

PhD student


work +49 6151 16-57 238
fax +49 6151 16-57 241

Work S3/19 1.3
Rundeturmstr. 12
64283 Darmstadt

Kai Cui joined the Lab as a PhD student in December 2019 and is part of the emergenCITY project funded by Hesse's research promotion programme LOEWE, sub-project KOM2. There, he is interested in applying reinforcement learning and mean field methods to optimize network configurations and drone communication networks.

Prior to that, he received his M.Sc. degree in Computer Science and in Electrical Engineering and Information Technology with focus on automation systems from Technische Universität Darmstadt in 2019.

His research topics includes reinforcement learning, game theory and mean field games. In particular, he is interested in new, scalable solutions for multi-agent problems. Possible applications for student projects could include learning drone swarm behaviour such as indoor exploration, or analyzing vaccination strategies.


Cui, K. ; Tahir, A. ; Sinzger, M. ; Koeppl, H. (2021):
Discrete-Time Mean Field Control with Environment States.
Preprint, 60th Conference on Decision and Control (CDC2021), Austin, USA, 13.-15.12.2021, [Conference or Workshop Item]

Cui, Kai ; Koeppl, Heinz (2021):
Approximately Solving Mean Field Games via Entropy-Regularized Deep Reinforcement Learning.
24th International Conference on Artificial Intelligence and Statistics, Virtual Conference, 13.-15.04.2021, [Conference or Workshop Item]

Ourari, R. ; Cui, K. ; Koeppl, H. (2021):
Decentralized Swarm Collision Avoidance for Quadrotors via End-to-End Reinforcement Learning.
Preprint, arXiv preprint, [Conference or Workshop Item]

Student Projects

Lalith Kishore Sai Macherla Sathish
(Master thesis)
Hierarchical Reinforcement Learning for Object Transportation via Morphological Drone Swarms running
Oliver-Maximilian Klein
(Bachelor thesis)
Reinforcement Learning for Mean Field Control with Environmental States running
Ruikang Sun
(Master thesis)
Estimation of Mean Field Models and Bayesian Reinforcement Learning running
Dongxu Wang
(Master thesis)
Multi-agent Reinforcement Learning and Consensus Algorithms running
Simon Schwab
(Bachelor thesis)
Morphological Control for Mid-Air Self-Assembled Drone Structures running
Davi Klein Spindola
(Bachelor thesis)
End-to-End Reinforcement Learning for Low-Level Control of Quadcopters running
Alexandre Knihar
(Bachelor thesis)
Depth Estimation and 3D Reconstruction for Micro Aerial Vehicles 06/2021
Tizian Claus Dege
(Bachelor thesis)
Neural Face Detection on Micro Aerial Vehicles 04/2021
David Askari-Badouee
(Bachelor thesis)
Decentralised Assembly of Drone Morphologies for Large Object Transportation 03/2021
Berk Calabakan
(Bachelor thesis)
Mid-Air Self-Assembling Drone Swarms 03/2021
Jiean Yan
(Master thesis)
Performance Comparison of Quadcopter Collision Avoidance Algorithms 01/2021
Shouran Mu
(Master thesis)
Potential and Rotational Flows for Collision-Free Drone Swarm Maneuvers 12/2020
Ramzi Ourari
(Master thesis)
Multiagent Collision Avoidance in Drone Swarms via Deep Reinforcement Learning 11/2020
Hongjing Tang
(Master thesis)
Low-Level Control of Quadcopters via Reinforcement and Imitation Learning 11/2020
João Victor Stahl Mosz
(Master thesis)
Performance Evaluation of Different Low-Level Quadcopter Control Algorithms 09/2020

Possible Student Projects you can find here.