A Trajectory Modification Method for Tool Operation Based on Human Demonstration Using MITATE Technique
Satonori Demura, Yaqiang Mo, Kotaro Nagahama, Kimitoshi Yamazaki
- Year
- 2018
- Citations
- 3
Abstract
We propose a novel method for a daily assistive robot to learn a new task from its user who does not have technical knowledge. “MITATE;' which is a human capacity to communicate a motion of a tool using a different tool and environment, provided inspiration to this system. The proposed system automatically estimates the mapping from the task space of the user's to the robot's. This enables the online teaching of a task by showing the target motion with a different tool and a different environment from the robot. We evaluated the proposed method with daily tasks experimentally, and the results proved that the proposed method is effective.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991