Hiroki Kanayama
Papers
2
Total Citations
17
H-Index
2
About
Hiroki Kanayama is a roboticist whose research focuses on advancing autonomous navigation and physical human-robot interaction. His key contributions lie in two areas: mapless visual navigation for mobile robots and data-driven control for humanoid robots. In his most-cited work, "Two-mode Mapless Visual Navigation of Indoor Autonomous Mobile Robot using Deep Convolutional Neural Network" (2020, 13 citations), Kanayama proposed a novel approach that eliminates the need for environment maps—traditionally requiring significant effort to create—by using deep convolutional neural networks to enable robots to navigate visually. This work addresses the fundamental challenges of self-localization and path planning without map dependency. His second notable paper, "A data-driven approach to probabilistic impedance control for humanoid robots" (2019, 4 citations), presents an innovative method for synthesizing whole-body motions that integrate visual perception and reaction force, allowing humanoid robots to maintain stable physical interactions with their environments. By encoding behaviors that combine motion, force, and vision, Kanayama's work pushes toward more adaptive and intuitive robots. Though early in his career, his research demonstrates a clear trajectory toward creating robots that can navigate and interact with the world more naturally and autonomously.
Research Focus
Key Achievements
Top Papers
- 1
- 2