Riku Murai
Papers
6
Total Citations
52
H-Index
3
About
Riku Murai is a robotics researcher whose work spans two interconnected frontiers: intelligent path planning and ultra-efficient focal-plane vision processing. He is perhaps best known for developing PathBench, a benchmarking platform that systematically evaluates both classical and learning-based path planning algorithms under a unified interface — a contribution that has attracted nearly 40 citations across its publications and addressed a longstanding gap in mobile robotics methodology. His subsequent systematic comparison study further solidified PathBench as a practical tool for the robotics community navigating the transition from traditional algorithms like RRT and wavefront planners toward deep learning-based approaches. Alongside this, Murai has pioneered the application of Focal-Plane Sensor-Processors (FPSPs) to robotic perception, demonstrating that these highly resource-constrained, pixel-level computing devices can support demanding tasks including CNN inference, visual odometry, and high-frame-rate homography estimation. His BIT-VIO system exemplifies this vision, fusing focal-plane binary features with inertial data to achieve agile, low-power navigation. Collectively, Murai's research addresses a fundamental challenge in robotics: enabling fast, efficient scene understanding without sacrificing computational practicality — making his work highly relevant to the emerging field of edge robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2Systematic comparison of path planning algorithms using PathBench12 citations · 2022
- 3Compiling CNNs with Cain: focal-plane processing for robot navigation4 citations · 2022
- 4Visual Inertial Odometry using Focal Plane Binary Features (BIT-VIO)3 citations · 2024
- 5
- 6