M. Komura

About

M. Komura is a researcher in robotics and artificial intelligence, with a primary focus on decision-making under uncertainty and state estimation. Their most cited work, "Real-Time Decision Making with State-Value Function under Uncertainty of State Estimation – Evaluation with Local Maxima and Discontinuity" (2006, 5 citations), introduces the real-time QMDP method, a novel approach that enables robots to make optimal decisions even when their perception of the environment is imperfect. By integrating a particle filter for state estimation with a state-value function derived from dynamic programming, Komura’s method evaluates all possible actions to select the best one in real time. This work addresses critical challenges such as local maxima and discontinuities in decision spaces, advancing the field of autonomous robotics. While their citation count is modest, the contribution is notable for its technical rigor and practical relevance to robotic navigation and control. Komura’s research continues to influence studies on robust, real-time decision-making in uncertain environments, making their work a valuable reference for students and researchers exploring the intersection of probabilistic estimation and reinforcement learning.

Research Focus

Key Achievements

Top Papers

1. 1
   Real-Time Decision Making with State-Value Function under Uncertainty of State Estimation – Evaluation with Local Maxima and Discontinuity

   5 citations · 2006

Key Collaborators

Contact & Links