About

Oliver Obst is a researcher whose work spans artificial intelligence, autonomous agents, and robotic systems, with particular emphasis on multi-agent coordination, robotic soccer, and machine learning for robot control. His career reflects a sustained commitment to bridging formal theoretical frameworks with practical robotic applications. Obst's early contributions focused on rational agent design, most notably his 2002 work on specifying agents using statecharts and utility functions (32 citations), which provided a structured approach to encoding agent decision-making. He extended this into multi-agent team coordination through hierarchical task network (HTN) planning (25 citations) and developed formal theoretical approaches to soccer strategy from behavioral specifications (25 citations). His 2004 research on model-based diagnosis for spatial reasoning (20 citations) demonstrated breadth beyond robotics into knowledge representation. A distinctive thread in his later research involves musculoskeletal robot control, where he applied Echo State Gaussian Process Regression to real-time inverse dynamics learning (15 citations), tackling the notoriously difficult problem of controlling artificial muscle actuators. His sustained involvement with RoboCup—spanning competitions, simulation leagues, and methodology—underscores his influence on reproducible robotic research infrastructure. Across these contributions, Obst has helped shape how autonomous robotic teams are designed, coordinated, and rigorously evaluated.

Research Focus

Key Achievements

8
H-Index
16
Papers
197
Total Citations
12
Avg Citations/Paper
🏆 Most Cited Paper
Specifying Rational Agents with Statecharts and Utility Functions
32 citations · 2002
📈 Most Prolific Year: 2018 (2 Papers)
🤝 Key Collaborators: 38
🏛 Institutions: University of Koblenz and Landau, University of Newcastle Australia, Commonwealth Scientific and Industrial Research Organisation, University of Bremen, Health Sciences and Nutrition, Western Sydney University

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