Miriam Zacksenhouse
Papers
14
Total Citations
191
H-Index
7
About
Miriam Zacksenhouse is a leading researcher in robotic control, specializing in the intersection of oscillatory neural networks, impedance control, and reinforcement learning for dynamic and contact-rich tasks. Her pioneering work on robotic yo-yo playing—a challenging, open-loop unstable periodic task—demonstrated how coupled oscillators can achieve phase-locked stabilization, with her foundational paper on oscillatory neural networks for yo-yo control accumulating 52 citations. She has made significant contributions to assembly robotics, notably developing reinforcement learning frameworks for impedance policies, including her 2022 paper on asymmetric matrices for peg-in-hole tasks (42 citations), and introducing residual admittance policies for learning contact-rich skills (15 citations). Her innovative "instantaneous model impedance control" method (15 citations) enhances robot-environment interaction by leveraging position controller error-correction. Zacksenhouse has also advanced bio-inspired locomotion, designing open-loop controllers based on central pattern generators for dynamic biped walking and minimal feedback strategies for slope-adaptive gaits. Her work bridges theoretical neural control principles with practical robotic dexterity, addressing fundamental challenges in unstable periodic motion, assembly automation, and adaptive locomotion.
Research Focus
Key Achievements
Top Papers
- 1Oscillatory neural networks for robotic yo-yo control52 citations · 2003
- 2
- 3Robotic Yoyo Playing With Visual Feedback16 citations · 2004
- 4Learning Contact-Rich Assembly Skills Using Residual Admittance Policy15 citations · 2021
- 5Instantaneous model impedance control for robots15 citations · 2002
- 6
- 7
- 8Return Maps, Parameterization, and Cycle-Wise Planning of Yo-Yo Playing7 citations · 2009
- 9
- 10