Tetsugaku Okamoto
Papers
2
Total Citations
4
H-Index
2
About
Tetsugaku Okamoto is a robotics researcher specializing in dynamic manipulation and trajectory planning, with a focus on integrating machine learning into control systems. His work addresses the challenge of enabling robots to perform complex, dynamic tasks—such as catching or throwing objects—by leveraging deep neural networks and sequence-to-sequence models. In his 2017 paper, Okamoto introduced a novel method combining model predictive control with deep neural networks, where target positions and model parameters are fed as inputs to generate optimal trajectories for dynamic manipulation. This approach has garnered 2 citations, highlighting its niche but foundational impact. His 2018 study further advanced the field by employing a hierarchical decoder in a sequence-to-sequence framework to plan trajectories with variable-length chunks, tackling the dynamic constraints inherent in robot manipulation. Though his citation counts are modest, Okamoto’s contributions are notable for their innovative fusion of control theory and deep learning, offering a pathway toward more adaptive and agile robotic systems. His work is particularly relevant for researchers exploring real-time motion planning in unstructured environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2