Papers
4
Total Citations
69
H-Index
4
About
Dr. Ayaka Kume is a pioneering roboticist whose work bridges the frontiers of autonomous underwater exploration and intelligent manipulation. Her research centers on developing robotic systems that can perceive, learn from, and adapt to complex, unstructured environments—from the deep sea to the warehouse floor. Dr. Kume’s major contributions include creating the first autonomous robotic surveys to volumetrically map tubeworm colonies in Kagoshima Bay, a breakthrough that automated the analysis of vast oceanographic datasets. Her foundational papers on this topic (2010, 2011) have garnered over 40 citations, establishing a new paradigm for efficient, data-driven marine biology. In parallel, she has made significant strides in robotic manipulation and reinforcement learning. Her work on end-to-end learning for object grasp poses, presented at the Amazon Robotics Challenge (2020, 20 citations), demonstrated how deep learning can enable robots to handle novel objects in cluttered settings. Additionally, her innovative Map-based Multi-Policy Reinforcement Learning framework (2017) allows robots to rapidly adapt to environmental changes or physical damage, a critical capability for mission-critical tasks. Dr. Kume’s research is distinguished by its practical impact, advancing both autonomous underwater vehicles and adaptive manipulation systems for real-world deployment.
Research Focus
Key Achievements
Top Papers
- 1
- 2End-to-End Learning of Object Grasp Poses in the Amazon Robotics Challenge20 citations · 2020
- 3
- 4