2007 Fiscal Year Final Research Report Summary
Construction of Pain Model and its Application to Safety Control
| Project/Area Number |
18560252
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent mechanics/Mechanical systems
|
|---|
| Research Institution | Kumamoto University |
|---|
Principal Investigator |
MATSUNAGA Nobutomo Kumamoto University, Graduate School of Science and Technology, Associate Professor (10363508)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
KAWAJI Shigeyasu Kumamoto University, Graduate School of Science and Technology, Professor (30040421)
|
|---|
| Project Period (FY) |
2006 – 2007
|
| Keywords | pain model / mechanical pain / skin model / pain sensor / safety control / FEM analysis / SED |
|---|
| Research Abstract |
Safety concept has been progressed in factory automation fields. The principle of basic safety is to separate the human from robot in the sense of space and time. However, according to the advance of robot into home space, it will be difficult to separate them. One approach to realize the safe coexistence is to make the physical regulations or limits to the output power, working velocity, size or weight of robots and so on. Other approach is to control the robot with the human-like pain feeling. It is well known that the learning ability of human is improved by causal linkage between action and subjective feeling. Therefore, embedding subjective feeling into robot controller will be a key of the new safety concept.
In this research, we have investigated about the pain generation mechanism considering the skin dynamics. And the results are summarized as follows :
1. Construction of the pain model
Most of the pain caused by physical interaction between robots and humans is the mechanical pain. A novel artificial superficial pain model (ASPM) caused by impact was proposed focusing the I/O relation between the force and subjective pain level. This ASPM model consists of mechanical pain model using two mass systems, skin model using elastic model and artificial neural networks based on gate control theory.
2. Parameters of pain characteristics
Some pain model parameters of upper limb were identified.
3. Safe control system using pain model
The ASPM is applied to avoidance motion control using virtual internal model. The effectiveness of the proposed model was evaluated by simulations using DD arm robot.
4. Prototype pain sensors
The prototype of the pain sensor simulated the laminated of the skin is developed. A new flexible pain sensor constructed by three layers, i.e. epidermis, fat and muscle, has been developed. The output of sensor was evaluated by FEM analysis and experiments.
|
