About

Keyan Ghazi-Zahedi is a leading researcher in embodied artificial intelligence and morphological computation, whose work explores how the physical body—not just the brain—contributes to intelligent behavior. His most influential paper, "SO(2)-Networks as Neural Oscillators" (92 citations), established foundational principles for using recurrent neural networks to generate rhythmic control signals in robots. He has made seminal contributions to understanding morphological computation, notably through his papers "Evaluating morphological computation in muscle and dc-motor driven models of hopping movements" (24 citations) and "Morphological computation: The good, the bad, and the ugly" (20 citations), which clarify how body softness, compliance, and dynamics can offload computational tasks from the controller. Ghazi-Zahedi also developed YARS, a physical 3D simulator for evolving robot controllers (18 citations), and demonstrated how small evolved neural networks can achieve complex behaviors like quadrupedal locomotion and obstacle avoidance. His work on information maximization in the sensorimotor loop (7 citations) introduced a novel, assumption-free learning method for self-organizing embodied systems. Through these contributions, Ghazi-Zahedi has shaped modern understanding of how morphology and control co-determine adaptive behavior in robots and biological systems.

Research Focus

Key Achievements

7
H-Index
10
Papers
193
Total Citations
19
Avg Citations/Paper
🏆 Most Cited Paper
SO(2)-Networks as Neural Oscillators
92 citations · 2003
📈 Most Prolific Year: 2003 (3 Papers)
🤝 Key Collaborators: 16
🏛 Institutions: Fraunhofer Institute for Intelligent Analysis and Information Systems, Max Planck Institute for Mathematics in the Sciences

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 · 13 days ago