Papers
3
Total Citations
66
H-Index
2
About
Pooja Guhan’s research lies at the intersection of robotics, deep reinforcement learning, and humanoid motion control, with a primary focus on enabling stable, real-time manipulation in complex robotic systems. Her most influential work introduces a deep reinforcement learning framework for solving the dynamically stable inverse kinematics of humanoid robots—a critical challenge given these robots’ inherent instability. By generating joint-space trajectories that maintain balance during motion, her approach offers a faster, more adaptive alternative to traditional analytical methods, directly addressing the need for real-time control. This foundational paper has garnered over 58 citations, underscoring its impact on the field. Guhan has also advanced dual-arm coordination in humanoids with articulated torsos, proposing a learning-based strategy that simplifies complex planning for reachability tasks, making online motion planning more feasible. Her contributions are particularly notable for bridging the gap between reinforcement learning and practical humanoid robotics, offering scalable solutions for dynamic environments. Through her work, Guhan has established herself as a key contributor to the development of more agile, stable, and capable humanoid robots, with implications for assistive robotics, autonomous navigation, and human-robot interaction.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3