Kanta Watanabe

Papers

Total Citations

H-Index

About

Kanta Watanabe is a researcher in developmental robotics and motor learning, with a focus on how artificial agents can autonomously acquire motor skills through self-exploration. Their key research areas include motor babbling, exploitation-exploration trade-offs in learning, and adaptive control for robotic manipulators. Watanabe’s most notable contribution is the introduction of “epsilon-greedy babbling” (2017, 5 citations), a novel framework that balances random exploration with targeted exploitation during trajectory sampling, enabling robots to learn drawing tasks without prior knowledge of their own body dynamics. This work bridges reinforcement learning principles with developmental robotics, offering a computationally efficient alternative to purely random babbling. In related work, Watanabe demonstrated fully automated learning for position and contact force control using wired flexible finger joints (2016, 3 citations), showcasing practical applications in soft robotics. Their 2018 study on the exploration-exploitation balance in motor babbling (2 citations) further refines these ideas. Though early in their career, Watanabe’s contributions are shaping how robots can autonomously learn from their environment, with potential impacts on assistive robotics and autonomous manipulation.