Papers
10
Total Citations
193
H-Index
7
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
Top Papers
- 1SO(2)-Networks as Neural Oscillators92 citations · 2003
- 2
- 3Morphological computation: The good, the bad, and the ugly20 citations · 2017
- 4YARS: A Physical 3D Simulator for Evolving Controllers for Real Robots18 citations · 2008
- 5Evolved Neurodynamics for Robot Control8 citations · 2003
- 6
- 7AN EVOLVED NEURAL NETWORK FOR FAST QUADRUPEDAL LOCOMOTION7 citations · 2007
- 8Adaptive Behavior Control with Self-regulating Neurons7 citations · 2007
- 9Editorial: Recent Trends in Morphological Computation5 citations · 2021
- 10