MohammadHossein Koosheshi
Papers
2
Total Citations
7
H-Index
2
About
MohammadHossein Koosheshi is a rising researcher in the field of robotic manipulation, with a primary focus on grasp detection and scene understanding. His work addresses the fundamental challenge of enabling robots to interact with their environment as adeptly as humans. In his highly cited 2023 paper, "AGILE: Approach-based Grasp Inference Learned from Element Decomposition," Koosheshi introduced a novel method that allows a robot manipulator to learn optimal grasping strategies by analyzing hand-object positioning, moving beyond simple object recognition to consider the approach path itself. Building on this, his 2024 work, "Multi-Modal Robust Geometry Primitive Shape Scene Abstraction for Grasp Detection," tackles the critical issue of scene understanding in unknown environments. By simplifying complex scenes into predefined geometric primitives, his research enables robots to perform more robust and reliable grasping tasks, bridging the gap between simulation and real-world application. With a cumulative citation count reflecting the immediate relevance of his contributions, Koosheshi is establishing himself as a key voice in the next generation of robotic intelligence, pushing the boundaries of how machines perceive and physically interact with the world.
Research Focus
Key Achievements
Top Papers
- 1AGILE: Approach-based Grasp Inference Learned from Element Decomposition5 citations · 2023
- 2