Sunin Kim

Papers

Total Citations

H-Index

About

Sunin Kim is a robotics researcher whose work focuses on advancing autonomous manipulation through reinforcement learning and skill discovery. Their key contributions lie at the intersection of computer vision and robotic control, particularly in developing systems that enable robots to learn complex object manipulation tasks from visual inputs. Kim's most cited work, "Object manipulation system based on image-based reinforcement learning" (2022, 5 citations), demonstrates a pioneering approach to training robots to interact with objects using only camera data, bypassing the need for explicit programming. Building on this foundation, their 2023 paper "Safety-Aware Unsupervised Skill Discovery" (3 citations) addresses a critical challenge in modern robotics: how to program increasingly complex manipulation behaviors in dynamic, unstructured environments. This work introduces novel algorithms that allow robots to autonomously discover and learn a diverse repertoire of safe manipulation skills without human supervision. Kim's research is particularly notable for tackling the scalability problem in robotic learning, where the growing number and complexity of tasks often overwhelms traditional programming methods. Their contributions are helping pave the way toward more adaptable, autonomous robots capable of operating safely in real-world settings.