15.07.2014 Views

Conference Program of WCICA 2012

Conference Program of WCICA 2012

Conference Program of WCICA 2012

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

<strong>WCICA</strong> <strong>2012</strong><br />

Book <strong>of</strong> Abstracts<br />

Book <strong>of</strong> Abstracts<br />

Friday, July 6, <strong>2012</strong><br />

PL-1 8:30-9:30 Room 305<br />

Plenary Lecture I<br />

Chair: Xie, Lihua<br />

Nanyang Technological University, Singapore<br />

◮ PL-1 8:30-9:30<br />

Zone Model Predictive Control <strong>of</strong> an Artificial Pancreas<br />

Doyle, Francis<br />

University <strong>of</strong> California at Santa Barbara<br />

Type 1 diabetes mellitus (T1DM) is a chronic autoimmune disease affecting<br />

approximately 25 million individuals in the world, and is the 4th<br />

leading cause <strong>of</strong> global death by disease. Current treatment requires<br />

either multiple daily insulin injections or continuous subcutaneous (SC)<br />

insulin infusion (CSII) delivered via an insulin infusion pump. Both<br />

treatment modes necessitate frequent blood glucose measurements to<br />

determine the daily insulin requirements for maintaining near-normal<br />

blood glucose levels. More than 30 years ago, the idea <strong>of</strong> an artificial<br />

endocrine pancreas for patients with type 1 diabetes mellitus (T1DM)<br />

was envisioned. The closed-loop concept consisted <strong>of</strong> an insulin syringe,<br />

a blood glucose analyzer, and a transmitter. In the ensuing<br />

years, a number <strong>of</strong> theoretical research studies were performed with<br />

numerical simulations to demonstrate the relevance <strong>of</strong> advanced control<br />

design to the artificial pancreas, with delivery algorithms ranging<br />

from simple PID, to H-infinity, to model predictive control.<br />

Our algorithmic studies have focused on model predictive control, including<br />

safety constraints to prevent over-dosing, and multi-parametric<br />

implementation for regulatory review. Our latest work has focused on<br />

s<strong>of</strong>t output constraints using “zones” to emulate the medical outcome<br />

metrics. A recent extension <strong>of</strong> that work will be described in this talk,<br />

consisting <strong>of</strong> a multipartite zone model predictive controller (Multi-Zone-<br />

MPC).<br />

Multi-Zone-MPC provides different tunings for the MPC weights based<br />

on four regions <strong>of</strong> glycemia: hypoglycemia, normoglycemia, elevated g-<br />

lycemia, and hyperglycemia. Defining these four zones provides richer<br />

control tunings that result in safe and effective control.<br />

Our latest clinical investigations will be reviewed to demonstrate the<br />

medical-relevance <strong>of</strong> such an approach to a feedback-controlled artificial<br />

pancreas.<br />

This presentation is based on work coauthored with Eyal Dassau, Rebecca<br />

Harvey, Matt Percival, Benny Grosman, Howard Zisser, Dale Seborg,<br />

and Lois Jovanovic.<br />

PL-2 9:50-10:50 Room 305<br />

Plenary Lecture II<br />

Chair: Shen, Tielong<br />

Sophia University, Japan<br />

◮ PL-2 9:50-10:50<br />

Applying Model Predictive Control in Automotive<br />

Chen, Hong<br />

Jilin University, China<br />

The basis <strong>of</strong> model predictive control (MPC) is the on-line solution <strong>of</strong><br />

a constrained optimization problem updated by the actual state. The<br />

obtained control is injected into the system until the next sampling<br />

time, while the procedure is repeated whenever new measurements<br />

are available. Due to its ability to handle nonlinearity, to include various<br />

types <strong>of</strong> models predicting the future dynamics, to take time-domain<br />

constraints into account explicitly and to coordinate multiple performance<br />

requirements in the sense <strong>of</strong> optimization, MPC has become an<br />

attractive feedback strategy for designing control systems in automotive.<br />

The talk will discuss some aspects <strong>of</strong> applying MPC in automotive<br />

through some selected examples.<br />

PL-3 10:50-11:50 Room 305<br />

Plenary Lecture III<br />

Chair: Meng, Max, Q.-H.<br />

Chinese University <strong>of</strong> Hong Kong, China<br />

◮ PL-3 10:50-11:50<br />

Surgical Robotics: Different Successful Concepts in the Past and in<br />

Future<br />

Lueth, Tim C.<br />

Technical University <strong>of</strong> Munich, Germany<br />

Since 20 years, surgical navigation and robotics are two important technologies<br />

to improve the state <strong>of</strong> the art in medical treatment. To know<br />

where an instrument is located relative to a region <strong>of</strong> interest (organ,<br />

vessel, bone structure) inside <strong>of</strong> the body is <strong>of</strong> great importance to<br />

achieve a preplanned postoperative situation. To guide and to move<br />

an instrument by a robot is more complex but is definitely required for<br />

almost all kind <strong>of</strong> surgery.<br />

While surgical navigation became a standard in many medical disciplines,<br />

surgical robotics is still at it’s beginning. Todays great commercial<br />

success <strong>of</strong> only one company with just one robotics approach<br />

(Telemanipulation), should not mislead to an interpretation that surgical<br />

robotics is now successful. There are still more problems than solutions.<br />

Also the visibility <strong>of</strong> this company is not typical for medical device<br />

companies. In the talk, several navigation and robotics systems are p-<br />

resented, that were developed within Germany during the past 15 years<br />

with different success. All <strong>of</strong> them skipped the barrier from idea to clinical<br />

use to the market. A collection <strong>of</strong> videos shows the robots use.<br />

Nevertheless, there are different mechanisms that are important to consider<br />

if a medical robot should be successful. These mechanisms are<br />

discussed and also the rules for researcher to design robots as medical<br />

device from the very beginning. Also some in-between solutions such<br />

as “Navigated Control” are presented to explain when a robot is useful<br />

and why sometimes a different solution is more successful.<br />

In future we will see, surgical robots and medical instruments that are<br />

patient specific printed on demand based on generative manufacturing<br />

methods such as Selective Laser Sintering <strong>of</strong> biocompatible materials.<br />

FrA01 13:30–15:30 Room 203A<br />

Intelligent Control and Automation (I)<br />

Chair: Wang, Peijin<br />

Co-Chair: Dang, Zhaohui<br />

Yantai Univ.<br />

National Univ. <strong>of</strong> Defense Tech.<br />

◮ FrA01-1 13:30–13:50<br />

Development <strong>of</strong> a Reconfigurable Robot’s Turning Method with Line<br />

Configuration, pp.61–66<br />

Chang, Jian<br />

Wu, Chengdong<br />

Shang, Hong<br />

Li, Bin<br />

Shenyang Inst. <strong>of</strong> Automation (SIA), Chinese<br />

Acad. <strong>of</strong> Sci.<br />

northeastern Univ.<br />

organization<br />

Shenyang Inst. <strong>of</strong> Automation, Chinese Acad. <strong>of</strong><br />

Sci.<br />

The shape-shifting robot is one <strong>of</strong> the kinds <strong>of</strong> the robot and it can<br />

change it’s configuration according to the environment. A method is<br />

proposed to solve the shortcomings <strong>of</strong> the traditional method <strong>of</strong> robot’s<br />

chain turning, which can shorten the time and radius <strong>of</strong> turning. The<br />

turning resistance moment can be also reduced. The mathematical<br />

model is built and the resistance torque and required force is computed<br />

By the analysis <strong>of</strong> experiment, the current <strong>of</strong> jaw motor does not exceed<br />

the limit load <strong>of</strong> the motor and it is more quick and smooth comparing to<br />

the link-turning. The validity and enforceability <strong>of</strong> the turning is proved<br />

by simulations and experiments.<br />

◮ FrA01-2 13:50–14:10<br />

91

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!