Atsushi Kanehira
Papers
8
Total Citations
175
H-Index
4
About
Atsushi Kanehira is a leading researcher at the intersection of robotics, artificial intelligence, and human-robot interaction, with a primary focus on enabling intuitive robot programming through natural language and visual demonstrations. His most impactful work centers on leveraging large language models (LLMs) and vision-language models (VLMs) for robotic control. In his highly cited 2023 paper, "ChatGPT Empowered Long-Step Robot Control in Various Environments" (92 citations), Kanehira pioneered a few-shot method for translating natural-language instructions into executable robot actions using customizable ChatGPT prompts. He extended this line of research with "GPT-4V(ision) for Robotics" (60 citations), introducing a pipeline that enables one-shot visual teaching by analyzing human demonstration videos to generate robot programs. Kanehira has also made significant contributions to the Learning-from-Observation (LfO) framework, developing systems that allow household robots to be programmed through few-shot human demonstrations without coding expertise. His work on task-sequencing simulators and interactive task encoding systems further advances the practical deployment of intelligent robots in everyday environments. With a growing citation impact exceeding 175 total citations, Kanehira's research is shaping the future of accessible, language-driven robot control.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5Learning-from-Observation System Considering Hardware-Level Reusability4 citations · 2022
- 6
- 7
- 8Interactive Task Encoding System for Learning-from-Observation2 citations · 2023