Papers
4
Total Citations
20
H-Index
3
About
Keegan Go is a roboticist whose research sits at the intersection of multi-agent coordination, contact-rich manipulation, and human-robot skill transfer. His work addresses fundamental challenges in enabling robots to operate safely and precisely in complex, real-world environments. Go’s most impactful contribution is **RoboBallet** (2025, 7 citations), which pioneers the use of graph neural networks and reinforcement learning to solve the intricate problem of collision-free task allocation and motion planning for multiple robots in shared workspaces—a critical need for modern manufacturing. He further advances robot dexterity with **GenCHiP** (2024, 5 citations), demonstrating how Large Language Models can generate policy code for high-precision tasks requiring nuanced force and contact reasoning, a domain previously considered out of reach for LLMs. Earlier in his career, Go explored human control strategies through haptics for six-degree-of-freedom tasks (2014, 5 citations) and designed an autonomous mobile olfactory robot for chemical source localization (2004, 3 citations). Spanning from foundational human skill analysis to cutting-edge AI-driven multirobot planning, Go’s work consistently pushes the boundaries of how robots perceive, coordinate, and physically interact with their world.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4