Thanpimon Buamanee
Papers
3
Total Citations
47
H-Index
3
About
Thanpimon Buamanee is an emerging robotics researcher whose work sits at the cutting edge of imitation learning and autonomous robotic manipulation. Specializing in the integration of bilateral control with advanced machine learning architectures, Buamanee has made significant contributions to how robots learn complex manipulation tasks from human demonstrations — a challenge central to modern robotics. Their most recognized work, "Bi-ACT" (2024, 25 citations), proposes a novel framework combining bilateral control-based imitation learning with the Action Chunking with Transformer (ACT) model, enabling more dexterous and responsive robot arm control. Building on this foundation, "ILBiT" (2024, 13 citations) extends the approach by incorporating both position and torque information, offering richer sensory feedback for learning fine manipulation skills. Their 2025 paper, "ALPHA-α and Bi-ACT Are All You Need" (9 citations), demonstrates that force and position information are critical for both unimanual and bimanual task execution — importantly achieved through accessible, low-cost systems. Collectively, Buamanee's research advances the democratization of robot learning, making sophisticated manipulation capabilities more practical and deployable in real-world, everyday environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3