Maxim Monastirsky

About

Maxim Monastirsky is a robotics researcher whose work bridges the gap between compliant manipulation and intelligent learning for complex robotic tasks. His primary research areas include haptic-aware manipulation, object insertion, and sample-efficient reinforcement learning for dynamic control. Monastirsky’s major contributions lie in developing algorithms that enable robots to handle spatial uncertainties using compliant hands, moving beyond rigid grippers. His 2022 paper on haptic-based and SE(3)-aware object insertion demonstrates how compliant hands can overcome grasp uncertainties, a critical challenge in industrial assembly, and has garnered 15 citations for its practical impact. In parallel, his work on learning to throw with decision transformers showcases a novel approach to dynamic tasks, using only a handful of samples to train effective throwing policies—a significant advance over traditional reinforcement learning methods that require extensive data. This paper also holds 15 citations, highlighting its relevance to both manipulation and learning communities. Monastirsky’s research is notable for its focus on real-world applicability, addressing fundamental uncertainties in robotic systems while pushing the boundaries of sample efficiency. His work is essential reading for students and researchers interested in compliant manipulation, haptics, and data-efficient robot learning.

Research Focus

Key Achievements

Top Papers

1. 1
   Haptic-Based and $SE(3)$-Aware Object Insertion Using Compliant Hands

   15 citations · 2022
2. 2
   Learning to Throw With a Handful of Samples Using Decision Transformers

   15 citations · 2022

Key Collaborators

Contact & Links