|
 Invited Talk II |
Evolvable and Emergent Neural Systems |
| Name |
Seong G. Kong |
| Affiliation |
The University of Tennessee U.S.A. |
| Abstract |
The advent of evolvable and emergent systems is in response to the recent surge of interest in biologically inspired computing models that are able to adapt to dynamic environments.
Artificial neural systems can benefit from various characteristics of living organisms such as modular organization, self-reproduction, self-repair, adaptability, and fault tolerance. Evolution and emergence are two fundamental concepts of biological systems. Many biological systems evolve and exhibit emergent behaviors driven by self-organization. Evolvable neural systems adopt biological evolution as a fundamental form of adaptation. Evolutionary algorithms have been successfully applied to design network architectures and to optimize connection weights. In emergent neural systems, biological developmental models are used as a self-reproduction mechanism. A neural system grows from simple to complex architecture as it develops. ¡®Global' intelligent behaviors emerge from ¡®local' interactions of small, simple constructing elements. The purpose of this talk is to stimulate interest in evolvable and emergent neural systems and to stimulate researchers from various disciplines to explore the challenges and advances of biologically inspired computing paradigm. |
|
|
|
| Invited Talk III |
The Trend and Prospect of LCD Glass Handling Robots |
| Name |
Dong-Il Kim |
| Affiliation |
Samsung Electronics , Korea |
| Abstract |
The current rapidly increasing demand for liquid crystal display (LCD) devices requires larger panels and continuous improvement in productivity to enable the device manufacturers to be competitive in the market. Increasing the size of the mother glass, which is the basic material for panel production, is critical for satisfying both requirements.
In the production of LCD devices, the robot is a key unit in that it handles (i.e. loads/unloads) the mother glasses between process equipment , such as chemical vapor deposition (CVD) , etching, and lithography. The larger the size of the mother glass, the larger the handling robots and process equipment required. In a seventh-generation LCD manufacturing line, the size of mother glass is up to 1870 ¢¥ 2200 mm so the size of the handling robot must be sufficiently large. Although the handling robot is designed to meet this constraint, the manufacturing line requires a minimum footprint and cycle time for the handling robot. The cycle time is very restricted by the structure of the handling robot.
To satisfy such requirements for the handling robot, both optimal design of robot mechanism and advanced control algorithms such as high-precision path interpolation for continuous motion and real-time static deflection compensation should be taken into consideration.
Here, we present seventh-generation LCD handling robot s of Samsung Electronics, where all of these have been commercialized . They have been applied to the seventh-generation LCD manufacturing line of Samsung Electronics. In addition, we also discuss the concept of the eighth -generation LCD handling robot and control methods. |
|
|
|
| Invited Talk IV |
Haptics for Surgery Training and Medical Robotics |
| Name |
Hannes Bleuler |
| Affiliation |
EPFL |
| Abstract |
Recently there has been a lot of activity in the field of virtual reality (VR) and imaging for surgery assistance. The "output" is graphical, usually 2-D, sometimes 3-D. However, there are still very few systems with haptic interface. Based on several years of experience mainly in the field of surgery training simulators with force feedback, this contribution will present two or three existing haptic device in the field of medical robotics and challenges to be addressed for further progress in the field. The first one of the systems realized concerns Mini-Invasive Surgery (MIS) in laparoscopy, the second one in Gynoecology (hysteroscopy) and the third one radiology (intervention by catheters and guide wires in blood vessels). All three systems are based on similar electronics and low-level SW, but the different tasks require very different dedicated hardware and adapted VR models. |
|
|
|
| Invited Talk V |
Ubiquitous Service Robotics |
| Name |
Hajime ASAMA |
| Affiliation |
University of Tokyo |
| Abstract |
RT (Robot Technology) is expected to magnify the IT industry, which is characterized by the real time operation in the physical environment. Especially, integration of RT with ubiquitous computation technology will make a big impact on service industry. In this talk, a concept of service media for ubiquitous service RT in the context of service engineering is introduced, which provides users with various services in a ubiquitous computing environment. Examples of ubiquitous RT devices called intelligent data carrier, and their application to user adaptive systems, rescue systems, and security systems are introduced.
|
|
|
|
| Invited Talk VI |
Road Following in an Unstructured Desert Environment using Monocular Color Vision as Applied to the DARPA Grand Challenge |
| Name |
Carl D. Crane III |
| Affiliation |
University of Florida |
| Abstract |
This paper describes the development of an autonomous ground vehicle that is being
developed to participate in the October 2005 DARPA Grand Challenge. The authors of this
paper are members of Team CIMAR which was one of twenty five teams selected by DARPA to participate in March 2004 in the inaugural competition to develop an autonomous vehicle that can navigate from near Los Angeles to near Las Vegas at speeds averaging twenty miles per hour. Most of the event was held on open terrain and trails in a rocky desert environment. This paper describes ongoing activities in preparation for the October 2005 event with emphasis placed on the utilization of color monocular vision to identify regions of smooth terrain that can be navigated at high speed.
|
|
|
|
|
