Debasmita Ghose

About

Debasmita Ghose is a researcher at the intersection of human-robot interaction, assistive robotics, and collaborative AI. Her work explores how robots can serve not just as tools, but as adaptive partners that teach, learn, and coach humans in real-world settings. She has made significant contributions to designing robots that improve human performance and safety—whether through an in-home robotic coach that helps people correct exercise mistakes (16 citations), or through interactive policy shaping techniques that enable robots to adapt to human collaborators in real time (11 citations). Ghose also advocates for a paradigm shift in robotics, arguing that robots should be built to both teach and learn from humans, a vision she laid out in a 2021 paper that has shaped discussions on reciprocal human-robot skill transfer. Earlier in her career, she developed vision-based obstacle detection and avoidance algorithms for mobile robots, including the VITAR platform, demonstrating a long-standing commitment to making robots perceptually aware and autonomous. Her work is notable for its human-centered focus, blending technical rigor with a deep understanding of how robots can empower people in everyday life.

Research Focus

Key Achievements

Top Papers

1. 1
   The Impact of an In-Home Co-Located Robotic Coach in Helping People Make Fewer Exercise Mistakes

   16 citations · 2022
2. 2
   Interactive Policy Shaping for Human-Robot Collaboration with Transparent Matrix Overlays

   11 citations · 2023
3. 3
   Why We Should Build Robots That Both Teach and Learn

   7 citations · 2021
4. 4
   Vision based obstacle detection using 3D HSV histograms

   6 citations · 2011
5. 5
   A Tracked Mobile Robot with Vision-based Obstacle Avoidance

   3 citations · 2007

Key Collaborators

Contact & Links