Ninad Khargonkar
Papers
5
Total Citations
37
H-Index
4
About
Ninad Khargonkar is an emerging robotics researcher whose work sits at the intersection of robotic manipulation, grasping, and computer vision. His research focuses on enabling robots to better perceive and interact with objects in unstructured environments — a fundamental challenge in advancing real-world robot autonomy. Khargonkar's most prominent contribution is **MultiGripperGrasp**, a landmark dataset containing over 30 million verified grasps across 11 gripper types and 345 objects, spanning simple parallel-jaw grippers to complex five-fingered dexterous hands. With 13 citations since its 2024 release, this work addresses a critical gap in generalizable robotic grasping research. Complementing this, his **NeuralGrasps** framework introduced neural implicit representations for encoding multi-hand grasps into a shared latent space, demonstrating his interest in scalable, data-driven manipulation learning. His vision-focused contributions — including a self-supervised system for unseen object instance segmentation through long-term robot interaction and the **RISeg** framework leveraging body frame-invariant features — reflect a commitment to making robots perceptually robust in novel environments. His **SceneReplica** benchmark further advances reproducibility in manipulation evaluation. Collectively accumulating nearly 40 citations, Khargonkar's work is building foundational infrastructure for the next generation of intelligent robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5