| Project/Area Number |
60460110
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
機械力学・制御工学
|
|---|
| Research Institution | University of Tokyo |
|---|
Principal Investigator |
YOSHIMOTO Kenichi University of Tokyo, Faculty of Engineering, 工学部, 教授 (10011074)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MIYAMOTO Yasuo University of Tokyo, Faculty of Engineering, 工学部, 助手 (50010864)
KOBAYASHI Hajime University of Tokyo, Faculty of Engineering, 工学部, 助手 (10010860)
|
|---|
| Project Period (FY) |
1985 – 1986
|
| Project Status |
Completed (Fiscal Year 1986)
|
| Budget Amount *help |
¥7,200,000 (Direct Cost: ¥7,200,000)
Fiscal Year 1986: ¥2,300,000 (Direct Cost: ¥2,300,000)
Fiscal Year 1985: ¥4,900,000 (Direct Cost: ¥4,900,000)
|
|---|
| Keywords | Flexible Manipulator / Flexible Link / Finite Dimensional Model / Optimal Type I Servomechanism / 分散階層制御系 |
|---|
| Research Abstract |
Conventionally, the positioning accuracy of robot manipulators has been improved by increasing the rigidity of the system. A high rigidity, however, results in an increase in mass which leads to deterioration in efficiency and high-speed behaviour. By developing a control strategy in which the flexibility of the link is taken into account, we can design a light manipulator with high positioning accuracy. This paper presents a real-time control system using hierarchical multi-processors with a state feedback for each joint. Using this flexible manipulator system, a precise control at high working speed is demonstrated. The outcome of this research is summerized as follows:
1. A simple mathematical model of the flexible manipulator which considers only the first mode of the link proved to be sufficiently accurate, whereas a precise model considering higher modes is extremely complicated and therefore not suitable for control purposes.
2. In linearing the equation, non-linear feedback was partially introduced. This gave better results than a simple linear expansion around the equilibrium point.
3. Dynamic feedback of link deformation proved to be an effective means of active damping.
4. Compensation of Coulomb's friction by feedforward was effective in improving positioning accuracy and damping the residual elastic vibration.
5. In order to achieve further precision in positioning and to compensate for changes in payload or friction, a type I servomechanism was implemented for each joint.
6. In the designing process, the control system was reduced by means of modal decomposition neglecting high-order vibration modes caused by interference between different modes. Nevertheless, the manipulator for this study had no trouble with respect to spillover because of its sufficient structural damping.
|
