Ebrahim Kouchaki

Papers

Total Citations

H-Index

About

Ebrahim Kouchaki’s research lies at the intersection of humanoid robotics and intelligent control systems, with a focus on achieving more natural, human-like locomotion. His early work on the effect of toe-joint bending on biped gait performance (2010, 9 citations) was foundational in identifying a critical yet often overlooked component of humanoid structure—the toe joint—and demonstrating its role in improving walking efficiency and stability. This contribution helped shift design priorities in the field toward more anatomically accurate robots. More recently, Kouchaki has advanced into machine learning-based control, as seen in his 2023 paper on balance control using deep reinforcement learning (3 citations). In this work, he introduces a hierarchical control architecture that combines actor-critic neural networks to optimize balance policies, offering a scalable approach to dynamic stability. While his citation counts are modest, the conceptual impact of his work is notable: bridging classical biomechanics with modern AI. Kouchaki’s research is particularly valuable for students and engineers seeking to understand how small structural details—like a toe joint—can unlock major improvements in robotic performance and how reinforcement learning can replace traditional controllers for complex tasks.