Takamasa Koshizen
Papers
7
Total Citations
35
H-Index
3
About
Takamasa Koshizen is a robotics researcher whose work centers on mobile robot localization, probabilistic sensor modeling, and autonomous systems. His most significant contribution is the development of the Gaussian Mixture Bayes with Regularized Expectation Maximization (GMB-REM) framework, a sophisticated probabilistic approach to tackling the twin challenges of modeling and reducing uncertainty in mobile robot position estimation. Through a series of interconnected publications spanning 2000 to 2003, Koshizen systematically advanced this framework — beginning with foundational sensor selection techniques, progressing through architectural refinements, and ultimately integrating sensor fusion capabilities that combine odometry and sonar data for more robust localization. His most cited work (2003, 10 citations) demonstrates how fusing multiple sensor modalities within the GMB-REM paradigm yields meaningful improvements in real-world positioning accuracy. Beyond localization, Koshizen also explored cognitive robotics, contributing to research on humanoid robots guided by complex kinematic features. While his citation counts remain modest, his body of work represents a coherent and methodical research program that helped lay groundwork for probabilistic approaches to robot self-localization — a field that has since become foundational in autonomous robotics and navigation research.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Sensor Selection by GMB-REM in Real Robot Position Estimation7 citations · 2000
- 4
- 5
- 6
- 7Cognitive Humanoid Robots Based on Complex Kinematic Features2 citations · 2003