Sambaran Ghosal
Papers
2
Total Citations
52
H-Index
2
About
Sambaran Ghosal is a robotics researcher whose work bridges the gap between visual perception and intelligent control, with a focus on making robots more adaptable and precise in real-world environments. His primary research areas include reinforcement learning for robotic manipulation, dynamics learning for mobile robot control, and the integration of egocentric and third-person visual perspectives. In his highly cited 2022 work, "Look Closer: Bridging Egocentric and Third-Person Views With Transformers for Robotic Manipulation" (47 citations), Ghosal tackles the challenge of precision-based manipulation tasks by using transformer architectures to fuse multiple visual viewpoints, enabling robots to learn fine-grained motor control from visual feedback—a significant step toward reducing the engineering burden of traditional robotic systems. More recently, his 2024 paper on "Hamiltonian Dynamics Learning from Point Cloud Observations for Nonholonomic Mobile Robot Control" (5 citations) addresses the critical need for adaptive navigation, proposing a data-driven approach that learns dynamics directly from 3D point cloud data, allowing mobile robots to adjust their control policies under changing operational conditions without relying on rigid hand-designed models. Ghosal’s work is notable for its practical impact on autonomous systems, combining theoretical rigor with real-world applicability.
Research Focus
Key Achievements
Top Papers
- 1
- 2