Oliver Kosak

University of Augsburg

Papers

10

Total Citations

99

H-Index

7

About

Oliver Kosak is a robotics and multi-agent systems researcher whose work centers on autonomous robot ensembles, swarm intelligence, and self-organizing systems. His research addresses one of modern robotics' most pressing challenges: enabling heterogeneous groups of robots to dynamically adapt, coordinate, and reconfigure themselves for diverse real-world applications, from industrial transport to search and rescue operations. Kosak's most influential contribution is his concept of "multipotent systems," introduced in his most-cited paper (33 citations), which elegantly combines planning, self-organization, and physical reconfiguration to create robot ensembles capable of flexibly responding to changing mission demands. Rather than designing robots for narrow, specialized tasks, his framework allows aerial and ground robots to synthesize new capabilities on the fly by sharing and reallocating hardware resources — a genuinely novel architectural paradigm. Complementing this, his work on decentralized coordination using the Jadex multi-agent framework and his Maple-Swarm programming model demonstrates a consistent commitment to bridging theoretical foundations with practical implementation. His research on self-descriptive hardware devices further addresses the persistent "reality gap" between simulation and real-world deployment. With over 90 cumulative citations, Kosak has established himself as a thoughtful contributor to the growing field of adaptive, collective robotic systems.

Research Focus

Key Achievements

7
H-Index
10
Papers
99
Total Citations
10
Avg Citations/Paper
🏆 Most Cited Paper
Multipotent Systems: Combining Planning, Self-Organization, and Reconfiguration in Modular Robot Ensembles
33 citations · 2018
📈 Most Prolific Year: 2018 (3 Papers)
🤝 Key Collaborators: 16
🏛 Institutions: University of Augsburg

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago