About

Masashi Sugimoto is a robotics researcher focused on enabling autonomous decision-making in dynamic, unpredictable environments. His core research areas include robot control, reinforcement learning, state-action prediction, and multi-agent systems. Sugimoto’s major contribution lies in developing methods for real-time sequential decision-making, where robots predict future state-action pairs to adapt their behavior on the fly—a critical capability for machines operating in complex, real-world settings. His work on using online Support Vector Regression (SVR) for state-action prediction and flexible-weight coefficients has laid groundwork for more adaptive robot controllers. More recently, he has explored deep reinforcement learning architectures, including Dueling DQN and centralized critic approaches for multi-agent coordination, as well as tracking abilities of Deep Q-Networks using LiDAR-equipped mobile robots. While his citation counts (ranging from 1 to 8 per paper) reflect a niche but dedicated audience, his sustained research output from 2013 to 2024 demonstrates a consistent commitment to advancing autonomous robotics. Sugimoto’s work is particularly relevant for students and researchers interested in bridging classical control theory with modern deep RL for real-time robot adaptation.

Research Focus

Key Achievements

3
H-Index
9
Papers
33
Total Citations
4
Avg Citations/Paper
🏆 Most Cited Paper
The proposal for deciding effective action using prediction of internal robot state based on internal state and action
8 citations · 2013
📈 Most Prolific Year: 2015 (4 Papers)
🤝 Key Collaborators: 10
🏛 Institutions: Muroran Institute of Technology, National Institute of Technology, Tomakomai College, National Institute of Technology, Kagawa College, Ehime University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago