M. Asif Rana
Papers
2
Total Citations
8
H-Index
2
About
M. Asif Rana is a roboticist whose research centers on learning for robot motion generation, particularly in multi-task and geometrically constrained environments. His major contribution lies in developing structured, end-to-end learning frameworks that enable robots to coordinate multiple objectives simultaneously. Rana’s most cited work, “Towards Coordinated Robot Motions: End-to-End Learning of Motion Policies on Transform Trees,” introduces policies inspired by Riemannian Motion Policies (RMPflow), allowing robots to learn from human demonstrations while respecting the geometric constraints of their kinematic chains. This approach bridges the gap between classical motion planning and modern imitation learning, offering a principled way to generate smooth, coordinated motions. With over 8 citations across related papers, his work is gaining traction in the robotics community for its practical impact on dexterous manipulation and human-robot collaboration. Rana’s research is notable for its elegant fusion of differential geometry and deep learning, providing a scalable path toward more capable and intuitive robot control.
Research Focus
Key Achievements
Top Papers
- 1
- 2