|
Origami Folding Robot - A Case Study of
Dextrous Hand Design and Skilled Motion Teaching| old_uid | 9780 |
|---|
| title | Origami Folding Robot - A Case Study of
Dextrous Hand Design and Skilled Motion Teaching |
|---|
| start_date | 2011/03/11 |
|---|
| schedule | 14h |
|---|
| online | no |
|---|
| location_info | pyramide ISIR (55/65) |
|---|
| summary | Dexterous manipulation by a robotic hand is a difficult problem involving (1) how to design a robot that has a capability to achieve a target task and (2) how to control the designed robot to actually conduct the task. In this talk, I will show our approaches to these problems, taking “Origami folding” as a target task. In the first part of my talk, I will introduce an origami-folding robot that can fold a “Tadpole”, a simple origami form but needs “squash folding”. The robot was designed based on our design philosophy, i.e., not blindly mimicking the appearance of the human hand but understanding the essence of the target task and designing a mechanism as simple as possible. Designing nice robot hardware is not enough to perform skilled tasks. We have to implement manipulation skill to the robot in some way so that the robot can cope with the variation of environmental properties. However, manipulation skill, which is a sort of implicit knowledge, is very difficult to be transferred even among humans. Skill transfer from human to robot is even more difficult because humans and robots have dissimilar bodies.
In the second part, I will introduce a novel method to synthesize a desired trajectory and sensory feedback control laws for robots based on the statistical characteristics of direct teaching data by a human. The proposed method was applied to the origami-folding robot and experimental results showed that the success
rate and the folding quality of ``Valley-fold'' were improved.
Finally, I will discuss a problem such as “what is the manipulation skill for robots?”. |
|---|
| responsibles | <not specified> |
|---|
| |
|