Ashwin Vangipuram
Papers
1
Total Citations
75
H-Index
1
About
Ashwin Vangipuram is a leading researcher in robot learning, with a primary focus on deformable object manipulation and visual model-based reinforcement learning. His work tackles one of robotics' hardest challenges: enabling machines to handle soft, shape-shifting materials like cloth, cables, and fluids. In his highly cited 2020 paper, "Learning Predictive Representations for Deformable Objects Using Contrastive Estimation" (75 citations), Vangipuram introduced a novel framework that jointly optimizes visual representation learning and dynamics modeling. By using contrastive estimation, his approach overcomes the traditional difficulty of learning plannable representations for objects with complex, high-dimensional state spaces—a breakthrough that makes model-based control practical for deformable objects. This work has been instrumental in advancing robots' ability to predict and manipulate non-rigid materials, with applications ranging from automated manufacturing to surgical assistance. Vangipuram's research sits at the intersection of computer vision, control theory, and deep learning, and his contributions have shaped how the field approaches representation learning for interactive environments. His work continues to inspire new methods in sample-efficient robot learning and embodied AI.
Research Focus
Key Achievements
Top Papers
- 1