Shir Kozlovsky
Papers
1
Total Citations
42
H-Index
1
About
Shir Kozlovsky is a leading researcher in robotic manipulation, with a focus on learning-based control for complex assembly tasks. Her work centers on reinforcement learning of impedance policies, particularly for peg-in-hole insertion—a fundamental challenge in industrial automation. Kozlovsky’s most-cited paper, “Reinforcement Learning of Impedance Policies for Peg-in-Hole Tasks: Role of Asymmetric Matrices” (2022, 42 citations), introduces a novel approach that leverages asymmetric impedance matrices to improve contact-rich manipulation. This contribution addresses a critical bottleneck in robotics: the need for precise, adaptive control in uncertain environments without extensive task-specific coding. By combining reinforcement learning with Cartesian impedance control, her research enables robots to learn compliant behaviors that are both robust and efficient. Kozlovsky’s work has significant implications for manufacturing, where reducing the need for structured environments and manual programming can accelerate automation. Her achievements highlight the potential of learning-based methods to bridge the gap between theoretical control and real-world application, making her a rising voice in the field of robotic assembly and physical human-robot interaction.
Research Focus
Key Achievements
Top Papers
- 1