Mycal Tucker

Massachusetts Institute of Technology, American Institute of Aeronautics and Astronautics

Papers

Total Citations

H-Index

About

Mycal Tucker is a researcher at the forefront of human-robot interaction and multi-agent coordination, with a focus on enabling robots to seamlessly integrate into teams with humans or other autonomous agents. His work tackles the fundamental challenge of shared understanding—how robots can learn the unspoken rules, conventions, and representations that make collaboration natural. In his highly cited paper, "Learning Unknown Groundings for Natural Language Interaction with Mobile Robots" (2019, 15 citations), Tucker pioneered methods for robots to infer the meaning of human language in real-world contexts. He further advanced the field with "Adversarially Guided Self-Play for Adopting Social Conventions" (2020, 6 citations), which introduced a novel framework for robots to autonomously learn and adopt arbitrary social norms—like driving on the right or left—by playing against themselves in adversarial settings. His work on "Latent Space Alignment Using Adversarially Guided Self-Play" (2022, 4 citations) extends this approach to aligning internal representations between agents, a critical step for effective information sharing in heterogeneous teams. Tucker’s research is shaping the future of collaborative robotics, offering scalable solutions for robots to become intuitive, convention-aware teammates.