1998 Fiscal Year Final Research Report Summary
Biologically-Inspired Machines Composed of Many Integrated Micromachines
| Project/Area Number |
09450104
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent mechanics/Mechanical systems
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
FUJITA Hiroyuki Institute of Industrial Science, Univ.of Tokyo, Professor, 生産技術研究所, 教授 (90134642)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TOSHIYOSHI Hiroshi Institute of Industrial Science, Univ.of Tokyo, Lecturer, 生産技術研究所, 講師 (50282603)
|
|---|
| Project Period (FY) |
1997 – 1998
|
| Keywords | micromachine / integrated circuit / micromachining / cellar automaton / decentralized control / evolutionary algorithmn / micro actuator |
|---|
| Research Abstract |
Biological systems and organs, which are composed of many simple elements, perform complicated functions by coordinating the behavior of elements. Cilia and the immune system composed of limphocites are such examples. Inspired by biological systems, we have investigated the distributed architecture of mechanical systems composed of many micromachines. The investigation focused in fabrication and decentralized control of such machines. The micromachine element has sensors, actuators and electronic circuits, all integrated together. We selected an intelligent cellular conveyor as the research target. Micro actuators arrayed on a planar substrate convey objects on them. Based on the information from sensors of its own and surrounding micromachines, the control circuit in a micromachine determines the motion of each actuator. We obtained the following resuts :
(l)We studied the control logic of the micromachine for the positioning and orientating task. The task should be completed by the gr
… More
oup work of microactuators that are controlled individually by the control logic. We extended two-dimensional cellular automata to handle sensor input and information obtained from neighboring micromachines and adapted them as the controller. The control rules were optimized by evolutionary algorithm ; selection of better rules depended on the simulated performance of each rule. The simulation model included dynamics of object motion and friction. Rules that were capable of positioning and orientating an object evolved in less than 100 generations.
(2)We also designed the control circuits for a sorting task. The system recognizes objects with different shapes and conveys them in different directions according to their shapes. Its subsystems, such as arrayed actuators, a distributed logic for shape recognition and a high-voltage driver for actuators were fabricated and operated. The system integration of those subsystems is underway.
(3)We developed hybrid integration technologies of chips with electronic circuits and chips with micromachined sensors and actuators. The technologies allow us to fabricate both chips separately and to increase the yield One method is called a silicon motherboard technology. The one edge of each chip is machined into micro connectors by deep dry etching. Chips are inserted into receptors that are machined in a bigger chip, the motherboard. Another technology is a stacking integration using micro holes through the chip. We confirmed adequate electrical and mechanical interconnection between the electric chip and the micromachine chip through the holes. Less
|
