Naoki Kojo
Papers
2
Total Citations
35
H-Index
2
About
Naoki Kojo is a pioneering researcher in cognitive robotics, whose work focuses on enabling humanoid robots to perceive, learn, and act in unstructured, real-world environments. His key research areas include sensorimotor learning, multimodal integration, and geometric symbol representation for robot intelligence. Kojo’s most influential contribution, "Situation Recognition and Behavior Induction based on Geometric Symbol Representation of Multimodal Sensorimotor Patterns" (2006, 25 citations), introduced a novel framework for memorizing, abstracting, and generating time-series sensorimotor data using recurrent neural networks. This work laid the foundation for robots to recognize situations and autonomously decide behaviors by representing complex sensory patterns as geometric symbols. In a notable practical achievement, Kojo demonstrated the "Realization of Trash Separation of Bottles and Cans for Humanoids using Eyes, Hands and Ears" (2007, 10 citations), where he integrated vision, force, and audio sensors to enable a humanoid to perform a challenging daily-life task—sorting recyclables—despite highly variable sensor inputs. This work exemplifies his commitment to bridging high-level cognition with robust physical interaction. Kojo’s research has been instrumental in advancing the field of developmental robotics, inspiring subsequent work on lifelong learning and adaptive behavior in autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2