Ukemi-SafeFall
Teaching a humanoid robot to fall like a martial artist: rolling to soften the impact, then getting back up on its own.
Anh Duc Tran
Founding Engineer @ Robotensor
I build robots and learning-based systems.
My work spans robot learning, locomotion, embedded systems, and hardware, especially systems that have to work on real robots.
Selected work
All projects →Project →
Reinforcement Learning for Quadruped Walking
Training a four-legged robot to walk smoothly across different terrain in simulation, then deploying the learned controller on a real Unitree Go1.
ForteGrip
A compact robotic gripper designed to handle both fragile and heavy objects while remaining highly backdrivable.
Teensy–Robstride Bridge
A hardware and software bridge that lets a PC reliably control multiple robot motors over Ethernet.
My projects have increasingly pulled me toward two questions:
- 01
How can robots stay safe when things go wrong?
- 02
How should robots be designed if learning from them is part of the goal?
Research
Ukemi-SafeFall: An Ukemi-inspired Injury-Aware Falling and Recovery
IEEE-RAS International Conference on Humanoid Robots (Humanoids) · Accepted, to be presented · Co-first author
ForteGrip: A Compact, High-Force, Backdrivable Parallel Gripper
Under review · Co-author
IEEE World AI IoT Congress (AIIoT)
Background
Master of Science in Computer Science
University of Massachusetts Amherst, USA
Erasmus+ Exchange Scholarship in Mechatronics Engineering
University of Limerick, Ireland
Bachelor of Engineering in Mechatronics Engineering
Ho Chi Minh City University of Technology, Vietnam National University, Vietnam
Outside the lab
Places I've lived, studied and traveled.
Drag to pan · pinch or + / − to zoom