Ekaterina Chaikovskaya
Papers
1
Total Citations
2
H-Index
1
About
Ekaterina Chaikovskaya is a roboticist whose research centers on bipedal locomotion and deep reinforcement learning for humanoid robots. Her most-cited work, "Benchmarking the Full-Order Model Optimization Based Imitation in the Humanoid Robot Reinforcement Learning Walk" (2023), systematically investigates how reference trajectories influence the learning of stable, efficient gaits. By comparing imitation-based methods with model-free approaches, she provides critical insights for control developers navigating the trade-offs between guided learning and autonomous exploration. This work has already garnered 2 citations, signaling its relevance to the growing field of robot motor skill acquisition. Chaikovskaya’s contributions help bridge the gap between classical robotics control and modern reinforcement learning, offering practical benchmarks that accelerate the development of more adaptive and robust humanoid robots. Her research is particularly valuable for students and engineers seeking to understand when and how to leverage reference trajectories in training walking policies—a foundational challenge in legged robotics.
Research Focus
Key Achievements
Top Papers
- 1