Control of Systems with MEMS Sensors and Actuators via Data Mining Techniques

desertcockatooΔιαχείριση Δεδομένων

20 Νοε 2013 (πριν από 3 χρόνια και 7 μήνες)

75 εμφανίσεις

Control of Systems with MEMS Sensors and Actuators via Data Mining Techniques


Wesley W. Chu*, Chih
-
Ming Ho**

*Computer Science Department

**
Department of Mechanical and Aerospace Engineering

University of California, Los Angeles


Contact Information

Wesle
y W. Chu, P.I.

Computer Science Department

University of California, Los Angeles

Los Angeles, Ca 90095

Phone: (310)825
-
2047

Fax: (310)825
-
7578

Email:
wwc@cs.ucla.edu


Chih
-
Ming Ho, Co
-
P.I.

Department of Mechanical an
d Aerospace Engineering


University of California, Los Angeles

Los Angeles, Ca 90095

Phone: (310)825
-
9993

Fax: (310)206
-
2302

Email:
chihming@ucla.edu


WWW Page

http://w
ww.cobase.cs.ucla.edu


Keywords

MEMS sensors and actuators

Dynamic control


Delta wing flight control

Temporal and spatial data mining



Project Award Information



Award Number: IIS
-
0097538



Duration: 08/31/2001


08/31/2004



Title: Control of Systems with M
EMS Sensors and Actuators via Data Mining Techniques


Project Summary

To process and interpret the vast amount of sensor information in determining the actuation schema to
control the system is an open problem. We use a novel data mining technique to deriv
e classification rules
and association rules from training datasets that consist of multivariate input and output variables. Based on
the system
-
operating environment, the best applicable rule can be selected to derive the actuation schema
that drives the
system to the desired state. The input
-
output relationship for delta wing aircraft is highly
non
-
linear, specifically the transfer function between the sensors and actuators is extremely complicated.
We plan to use the delta wing aircraft input/output MEMS

test bed samples to develop a scalable data
mining technique that discovers full input
-
output relationship under a wide range of conditions (dynamic,
temporal, spatial, etc).


This project leverages on our past data mining research of multivariate variabl
es training datasets, as well
as the available MEMS sensors and actuators for UAV (Unmanned Aerial Vehicle) application and wind
tunnel measurement facility to collect the training data. Our current research is on the design of
experiments to collect the d
ata for dynamic system behavior, extending the mining algorithm for
summarizing temporal rules, developing the rule selection strategy for actuation schema, and developing
the wind tunnel experiments to validate our research approach.


Goals, Objectives, a
nd Targeted Activities

The goal of our research is to discover non
-
linear relationships between the distributed sensor input and
actuation schema output by using data mining techniques. We have designed a set of experiments to collect
the data for dynamic
system behavior, and are extending the mining algorithms to handle massive amounts
of multivariate training data. We plan to develop methodology to generate temporal rules and strategy to
automatically select actuation rules for real time system control.


Indication of Success

We will discover rules for representing sensor and actuation relationships based on vast amounts of non
-
linear and temporal sensor input and physical output data. Such rules can be used for the flight control of
delta wing aircrafts u
nder dynamic movement.


GPRA Outcome Goals

1.

Discoveries at and across the frontier of science and engineering.



Data mining techniques to generate actuation rules from distributed multivariate sensor data

We are developing methodology to generate temporal
rules that capture and summarize the characteristics
of dynamic data, to handle massive amount of multivariate sensor data and generate rules to represent the
multivariate input and output relations, and to automatically select actuation rules for real tim
e system
control. These data mining techniques are instrumental for any application requiring real time dynamic
control based on continuously streaming data from massively parallel sensors, so the results should play a
important role in controlling micro f
abricated
-
based systems.

2.

Connections between discoveries and their use in service to society.



Develop techniques for dynamic system control

Together with the researchers at Department of Mechanical and Aerospace Engineering, we are developing
techniques
to automatically read information from massive distributed sensors, transfer the information to
actuation rules and select the most applicable rule to derive the actuation schema. This process can be used
in other real time control systems, such as environ
mental, medical, etc.


Project Reference



Z. Liu, W.W. Chu, A. Huang, C. Folk, C.M. Ho,
Mining Sequence Patterns from Wind Tunnel
Experimental Data for Flight Control.
PAKDD 2001: 270
-
281.



G. Giuffrida, W.W. Chu, D.M. Hanssens,
NOAH: An Algorithm for Mining

Classification Rules from
Datasets with Large Attribute Space.

In Proceedings of 12
th

International Conference on Extending
Database(EDBT), Konsta, Germany, March 2000.



Q. Zou, W.W. Chu, D. Johnson, H. Chiu,
A Pattern Decomposition Algorithm for Finding A
ll frequent
Patterns in Large Datasets.

ICDM2001: 673
-
674.



W.W. Chu, K. Chiang, C.C. Hsu, H. Yau,
An Error
-
based Conceptual Clustering Method for
Providing Approximate Query Answers.

Communications of the
ACM
, 39(13), December 1996.



W.W. Chu, Q. Chen, A. H
wang,
Query Answering via Cooperative Data Inference.

Journal of
Intelligent Information System 3, 57
-
87, 1994.



J. Han, J. Pei, Y. Yin,
Mining Frequent Patterns without Candidate Generation
. 2000 ACM SIGMOD
Intl. Conference on Management of Data.



C.M. Ho,
P.H. Huang, J. Lew, J.D. Mai, V. Lee, Y.C. Tai,
Intelligent System Capable of Sensing
-
Computing
-
Actuating
, Keynote Address, 4
th

Intl. Conference on Intelligent Materials, Society of Non
-
Traditional Technology. Tokyo, Japan, October 1998.



C.M. Ho, P.H. Huan
g, J.M. Yang, G.B. Lee, Y.C. Tai,
Active Flow Control by MicroSystems,
FLOWCON, Intl. Union of Theoretical and Applied Mechanics (IUTAM) Symposium on Mechanics
of Passive and Active Flow Control, Gottingen, Germany, Sept.1998. pp18
-
19.



T. Tsao, F. Jiang, R
.A. Miller, Y.C. Tai, B. Gupta, R. Goodman, S. Tung, C.M. Ho,
An Integrated
MEMS System for Turbulent Boundary Layer Control.
Technical Digest, 1997 Intl. Conf. On Solid
-
State Sensors and Actuators (Transducers’97), Chicago, IL, Vol.1, pp.315
-
318, June 16
-
19 (1997).



C. Liu, J. Huang, A. Zhu, F. Jiang, S. Tung, Y.C. Tai, C.M. Ho,
A Micromachined Flow Shear Stress
Sensor Based on Thermal Transfer Principles
, IEEE/ASME J. of Microelectromechanical Systems
(J.MEMS), 1999.



Lee, G.B., Chiang S., Tai, Y.C, Tsao, T
., Ho, C.M.,
Robust Vortex Control of a Delta Wing Using
Distributed MEMS Actuators.
Journal of Aircraft (2000).



Huang, A., Ho, C.M., Jiang, F., Tai, Y.C.,
MEMS Transducers for Aerodynamics


A Paradym Shift.
38
th

Aerospace Science meeting & Exhibit, AIAA
00
-
0249. Reno, NV.



Area Background

Data mining, dynamic control of macro
-
scale machine, distributed sensor


Area References



Z. Liu, W.W. Chu, A. Huang, C. Folk, C.M. Ho,
Mining Sequence Patterns from Wind Tunnel
Experimental Data for Flight Control.
PAKD
D 2001: 270
-
281.



C.M. Ho, Y.C. Tai,
Micro
-
Electro
-
Mechanical
-
Systems (MEMS) and Fluid Flows Annual Review of
Fluid Mechanics
.


1998 30:579
-
612



Potential Related Projects

There are several temporal spatial data mining projects (e.g. Christos Faloutsos, Ma
tthew O. Ward, Vassilis
J. Tsotras) in the IDM. They all can be leveraged on each other’s research.