The following half day and full day tutorials and workshop will be held on October 31, 2000.

TA-1(9am-Noon, Oct. 31,  at Kagawa Industrial Intelligence Center)
Robotics and Neuroscience (Half-day Tutorial)
Organizer: Prof. Angel P. del Pobil, (Universidad Jaume-I)

TP-2(******* Canceled ***********)
Evolutionary Robotics (Half-day Tutorial)
Organizer: Prof. Dario Floreano (Swiss Federal Institute of Technology)
If you had registrated TP-2 already, please contact Sigeki Sugano who is the one of Tutorial and Workshop Co-Chairs.
sugano@paradise.mech.waseda.ac.jp

WP-3(1pm-4pm, Oct. 31, at Kagawa Industrial Intelligence Center)
Integrated Micro Sensors and Actuators for Robotics and Automation Applications(Half-day Workshop)
Organizers: Prof. Wen J. Li (The Chinese University of Hong Kong), Prof. Richard Yeh (UC Berkeley), Prof. Kris S. J. Pister (UC Berkeley), and Prof. Isao Shimoyama (The University of Tokyo)

TAP-4(9am-4pm, Oct. 31, at Sun Messe Kagawa)
Sensing and Actuation Toward 21st Century (Full-day Tutorial)
Organizer: Prof. Makoto Kaneko (Hiroshima University)

WAP-5(9am-4pm, Oct. 31, at Sun Messe Kagawa)
Personal Robotics (Full-day Workshop)
Organizer: Prof. Paolo Dario (Scuola Superiore Sant'Anna)

  Shigeki Sugano
    Department of Mechanical Engineering, School of Science & Engineering,
    Waseda University
    3-4-1 Ookubo, Shinjuku, Tokyo, 169-8555 Japan
    Tel: +81-3-5286-3264; Fax: +81-3-5272-0948;
    Email: sugano@paradise.mech.waseda.ac.jp

  Ning Xi
    Department of Electrical and Computer Engineering, Michigan State University
    2120 Engineering Building, East Lansing, Michigan 48824-1226 USA
    Tel: +1-517-432-1925; Fax: +1-517-353-1980; Email: xin@egr.msu.edu

  Martin Buss
    Institute of Automatic Control Engineering, Technische Universitat Munchen
    D-80290 Munchen, Germany
    Tel: +49-89-2892-3404; Fax: +49-89-2892-8340; Email: M.Buss@ieee.org
　

STATEMENT OF OBJECTIVES AND BACKGROUND

The motivation of this tutorial can be summarized by rephrasing one of Wiener's statements: The roboticist need not have the skill to conduct a neurophysiological experiment or propose a model, but (s)he must have the skill to understand one, to criticize one, and to suggest one. The neuroscientist need not be able to build and program a robotic system, but (s)he must be able to grasp its neurophysiological significance and to tell the roboticist for what (s)he should look.

The purpose of this tutorial is to make the attendee aware of the benefits that a closer interplay between robotics and neuroscience may have in the future -- in other words: What can Robotics learn from Neuroscience? and, conversely, What can Neuroscience learn from Robotics?

The control of complex robotic systems requires the interaction of mechanisms for perception, sensorimotor coordination, and motor control. There have been a number of successful applications of neural computation to robot control, however, most of these connectionist approaches are based on control theory formalisms. Only a few have been largely inspired by neurophysiological data. Some current approaches focus on simple sensory-motor mappings for the production of behavior, while not being concerned about the underlying neural mechanisms, or issues involving planning or higher-level control. Most analytical approaches to state-of-the-art problems in robotics are biologically implausible, but drive much of the current research in robotics. Although there exist computational models of neural mechanisms --such as visuo-motor or eye-hand coordination--  with evident implications for intelligent robotics, very few have been used in actual robot implementations.

Much progress has been made toward understanding the neural mechanisms involved in walking, reaching, and grasping, but these mechanisms are complex and only partially understood. The study of these behaviors and their mechanisms in animals and robots may lead to fruitful insights in both directions. As we learn more about the neurophysiology of living beings, we will be able to build better robots and, conversely, the construction and programming of robots may provide new hypothesis for the study of neural mechanisms.

INTENDED AUDIENCE

Researchers in  mobile robotics, robot manipulation, reasoning and planning, and robot sensing. R&D Engineers who are interested in applying these technologies to industry. They will get exposed to the state-of-the-art robotics research based on Neurophysiology in leading robotics labs. We hope to attract a larger than usual percentage of younger minds (graduate students) to the tutorial. It is also useful for robotics researchers who would like to evaluate important issues for the near future.

COMPLETE LIST OF SPEAKERS AND AFFILIATIONS

Angel P. del Pobil, Jaume-I University, Spain
pobil@ieee.org
Andrew Fagg, University of Massachusetts, USA
fagg@cs.umass.edu

DETAILED LIST OF TOPICS

1.- Introduction.
Successful applications of neural learning in robot control will be presented to show that inspiration in neurophysiological data is often missing.
1.1 Inverse kinematics
1.2 Visual servoing
1.3 Fine Manipulation
1.4 Mobile robot navigation
1.5 Autonomous driving

2.- Behavior-based robotics.
This new paradigm in intelligent robotics will be analyzed from a neurobiological standpoint.
2.1 Primitive behaviors and emergent intelligence
2.2 Neural Networks embedded in a robot body
2.3 Hybrid robotic systems

3.- Spinal circuitry and arm dynamics.
This sections examines the role of spinal circuitry and the stretch reflex in arm dynamics, with their implications in control.
3.1 Relationship to P-D control
3.2 Dynamical implications of the nonlinearities introduced by the stretch reflex

4.- Learning in an embedded neural system.
The fundamental issues about learning in robots and animals will be addressed: e.g., incremental learning, focus-of-attention problem, etc.
4.1 Reinforcement Learning, and the basal ganglia
4.2 Active Learning and Exploration
4.3 Supervised Learning and the cerebellum

5.- Active Vision.
The role of vision in animals and robots seen as real-time perception-action systems.
5.1 Goal-oriented perception
5.2 Exploratory motions and image flow
5.3 Attentional mechanisms: saccades and selective visual attention

6.- Visuomotor coordination in primate heads.
Some mechanisms of visually-controlled behaviors will be presented in sections 6 and 7.
6.1 The vestibulo-ocular reflex
6.2 Visual stabilization and corollary discharge
6.3 Visual localization and remapping

7.- Visuomotor coordination in the fly.
The fly visual system will be described as a simple but robust and efficient example of visuomotor coordination.
7.1 Retinal motion patterns and motor activity
7.2 Information processing in the fly visual system

8.- Rhythmic behaviors.
The basic rhythmic behaviors involved in flight or locomotion will be considered as building blocks of more complex motor behaviors.
8.1 Motor pattern generators
8.2 Neural control of rhythmic robot arm movements

9.- Eye-hand coordination for reaching.
This section analyses reaching in robots and animals: i.e. visually guided movement to bring the hand toward an object location in space, presenting some controversial issues.
9.1 Analytical and connectionist approaches to robot reaching
9.2 Representation of space:
      explicit vs. distributed self-organizing oculomotor
      representation
9.3 Encoding goal-directed movement motor commands

10.- Cortical mechanisms in primate grasping.
Recent advances in the understanding of the neural mechanisms involved in grasping are contrasted with the current trends in robot grasping research.
10.1 Representation of objects and grasps in the cortex
10.2 Coupling between parietal and premotor cortex
10.3 Infant development of reaching/grasping skills
10.4 Current approaches to robot grasping

11.- Conclusion.
Summary of main lessons learned about the interplay between robotics and neuroscience.
 

SPEAKERS' BIOGRAPHICAL SKETCHES:

         Angel Pasqual del Pobil is Professor of Computer Science and Artificial Intelligence at Jaume I University (Spain) and director of the Robotic Intelligence Laboratory. He holds a B.S. in Physics (Electronics, 1986) and a Ph.D. in Engineering (Robotics, 1991), both from the University of Navarra. His Ph.D. Thesis was the winner of the 1992 Award of the Spanish Royal Academy of Doctors. He is Co-Chair of the Robot Motion & Path Planning Technical Committee of the IEEE Robotics and Automation Society and was Vice President of the International Society of Applied Intelligence (1996-1999). He is author or co-author of more than 75 scientific publications --including the book Spatial Representation and Motion Planning (Springer, 1995)-- and co-editor of three books: Practical Motion Planning in Robotics (John Wiley & Sons, 1998), and Springer LNCS/LNAI 1415 and 1416. Prof. del Pobil is co-organizer of workshops at the 1996 and 2000 IEEE International Conference on Robotics and Automation and he has been Program Co-Chair of the 11th International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems (IEA/AIE-98). He has served on the program committees of 23 international conferences, such as IEA/AIE (1997-2000) , IEEE ICRA (1999-2000), ICAR (1997), IEEE Int. Symp. on Computational Intelligence in Robotics and Automation (1999), Int. Workshop on Artificial and Natural Neural Networks (IWANN 1999), etc. He also serves as a technical reviewer for several international journals and conferences.
         He  has been involved in robotics research for the last fourteen years and has worked on different topics such as: motion planning and collision avoidance (mainly for robot arms), visually-guided grasping, sensorimotor transformations, self-organization in robot perception, neural and reinforcement learning for sensor-based manipulation, etc. He was invited as a plenary speaker on "Robotics and Neuroscience" at IWANN99 (Springer LNCS 1606). Professor del Pobil has fourteen years of teaching
experience (both  undergraduate and graduate). He has been speaker of several tutorials in international conferences held in Melbourne, Berlin, Leuven, etc.
http://titan.act.uji.es/pobil/

Andrew H. Fagg is a research assistant professor in the Department of Computer Science at the University of Massachusetts, Amherst.  He holds a B.S. (1989) in Applied Mathematics (Computer Science) from Carnegie-Mellon University, and a M.S. (1991) and a Ph.D. (1996) in Computer Science from the University of Southern California.  Research interests include learning mechanisms for motor control in biological (especially primate) and robotic systems, as well as machine learning and wearable computing.  Current work includes computational  models of learning in the cerebellum and the basal ganglia, with a focus on their involvement in reaching and grasping.
http://www.cs.umass.edu/~fagg/

CONTACT PERSON:

Angel P. del Pobil
Email: pobil@ieee.org
Robotic Intelligence Laboratory
Universidad Jaume-I               Tels.: +34-964-72.82.93 (office)

Campus Riu Sec, Edificio TI              +34-964-72.82.94 (Lab)
E-12080 Castellon                 Fax:   +34-964-72.84.35
SPAIN                             http://titan.act.uji.es/pobil/

(Back to the list)

Background and Objectives:

The progress of robotics technology and the growing social importance and attention in the industrialized countries for the expectations and needs of the disabled and elderly have determined the development of the field of research and industrial application known as "Personal Robotics", whose objective is to implement intelligent machines capable of assisting human beings needing help in many practical circumstances. This copes with two main objectives that have been recurrent in the history of the evolution of humankind for the "engineers" who designed machines: the first was the "dream" of developing human-like machines ("humanoids") which aimed at replicating human functions, shape and even reasoning; the second was the "need" for implementing useful machines which could help human beings in real-life conditions and alleviate their (often hard) working conditions.

The development of Personal Robots is being pursued by many research groups world-wide, by adopting different approaches in Japan, USA and Europe. The workshop, intended both for academic and industrial researchers in robotics, aims at bringing together the main researchers in this field, so as to present current achievements, compare different approaches, discuss the cultural and ethical implications, highlight the challenges to be addressed and finally state the potentiality of this research field.

Schedule:
 9.00 -  9.30: PAOLO DARIO: 'Introduction to Personal Robotics'
 9.30 - 10.00: KAZUHITO KAWAMURA
10.00 - 10.30: COFFEE BREAK
10.30 - 11.00: KAZUO TANIE
11.00 - 11.30: RUDIGER DILLMAN: 'Learning and cooperative robots in daily living environments'
11.30 - 12.00: MASAKATSU FUJIE: 'How to be familiar with our service robots for general elderly'
12.00 - 13.00: LUNCH
13.00 - 13.30: TAKANORI SHIBATA: 'Synthesis of art and technology for creation of subjective value on personal robot'
13.30 - 14.00: HENRIK CHRISTENSEN: 'Mobile manipulation for intelligent service robotics'
14.00 - 14.30: COFFEE BREAK
14.30 - 15.00: MASAHIRO FUJITA: 'Challenges in robot entertainment systems'
15.00 - 15.45: PANEL DISCUSSION (all speakers as panelists)
15.45 - 16.00: PAOLO DARIO: Conclusions

Speakers:

Masakatsu FUJIE
Mechanical Engineering Research Laboratory
Hitachi Ltd.
502 Kandatsu, Tsuchiura, Ibaraki, 300-0013 Japan
TEL +81 - 298-32-4111
FAX +81 - 298-32-2806
Email: mgfujie@merl.hitachi.co.jp

Masahiro FUJITA
Digital Creatures Laboratory
Sony Corporation
6-7-35 Kitashinagawa, Shinagawa-ku, Tokyo, 141-0001 Japan
Tel 03-5448-5901
Fax 03-5448-6833
Email: mfujita@pdp.crl.sony.co.jp

Kazuhito KAWAMURA
Electrical and Computer Engineering Dept.
Jacobs Hall, Rm 322
Box 1764 Station B
Vanderbilt University
Nashville, TN 37235
Tel. +1-615- 322-2735/2771
Fax: +1-615- 343-6702
E-mail: kawamura@vuse.vanderbilt.edu

Paolo DARIO
Advanced Robotics Technology and Systems Laboratory - ARTS Lab
Scuola Superiore Sant'Anna
Via Carducci 40
56127 Pisa, Italy
Tel. +39-050-883400
Fax: +39-050-883402
E-mail: dario@arts.sssup.it

Henrik CHRISTENSEN
Centre for Autonomous Systems, CVAP
Numerical Analysis and Computing Science
Kungliga Tekniska Hogskolan
S-100 44 Stockholm, Sweden
Phone: +46 8790 6792 (work)
Secretary: +46 8790 6227
Fax: +46 8723 0302
E-mail: hic@nada.kth.se

Takanori SHIBATA
Bio-Robotics Division
Department of Robotics
Mechanical Engineering Laboratory, AIST-MITI
1-2 Namiki, Tsukuba 305, Japan
Tel 0XXX-XX-7299
Fax 0XXX-XX-7201
Email: shibata@mel.go.jp

Rudiger DILLMAN
Department of Computer Science
Institute for Process Control and Robotics
University of Karlsruhe
Germany
Tel. 608-3846
Fax 608-7141
E-mail: dillmann@ira.uka.de

Topics:

Humanoid robotics
Health care and rehabilitation robotics
Service robots
Entertainment robots
Mobile manipulation
Human-robot interaction
Bio-robotics
Emotional behavior

(Back to the list)

Background and Objectives:

Evolutionary Robotics is a technique for the automatic creation of autonomous robots. Inspired upon the Darwinian principle of selective reproduction of the best individuals and the principles of genetics, Evolutionary Robotics looks at robots as autonomous artificial organisms that develop their skills in close interaction with the environment without human intervention. Heavily drawing from natural sciences like biology, cognitive science and ethnology, Evolutionary Robotics resorts to bio-inspired engineering techniques such as neural networks, genetic algorithms, dynamical systems, and bio-morphic engineering.

This tutorial will provide in-depth knowledge of the methodology, give practical hints as to how one might design a fitness function, suggest ways of combining evolution and learning, show the relevance of results for a new vision of Artificial Intelligence, and describe several examples of evolving robots. Finally, it will address issues related to the application of this technology for practical and commercial systems. The goal is to inform the audience and at the same time inspire new experiments and applications.

Speakers:

Dario Floreano
Laboratory of Microprocessors and Interfaces (LAMI)
Swiss Federal Institute of Technology, Lausanne (EPFL)
Switzerland
Dario.Floreano@epfl.ch

Stefano Nolfi
Division of Neural Systems and Artificial Life
Italian National Research Council (CNR), Roma
Italy
nolfi@ip.rm.cnr.it

Takashi Gomi
Applied AI Systems, Inc.
Ottawa, Canada
gomi@aai.ca

Detailed Plan:

1. Why Evolutionary Robotics
     Engineering motivation
     Artificial Life motivation
     Biological/Cognitive motivation
2. Background on artificial evolution
     Genetic Algorithms
     Artificial Neural Networks
     Interfacing GAs and ANNs
     Genetic Programming and other techniques
3. Methodological Issues
     Evolution in simulation and on physical robots
     The evaluation issue
     The design of the physical robot
4. Evolution of simple behaviors
     Straight navigation and obstacle avoidance
     Visually-guided navigation
      Re-adaptation in changing environments
      Evolution of walking
5. Evolution of sensory-motor intelligence
     Self-selection of input stimuli
     Tasks that apparently require memory or internal representations
     Active perception
6. Beyond sensory-motor intelligence
     Emergence of internal representations
     The role of modularity
     Dynamical controllers
7. Learning and Evolution
      The adaptive functions of learning in evolution
      How learning can 'help and guide' evolution
      Prediction learning and evolution
      Evolution of learning mechanisms
      Adaptation to fast changing environments
8. Competitive Co-Evolution
      Changing fitness landscapes
      Evolving Predator-prey robots
      Learning in competitive co-evolution
9. Evolvable Hardware
     Evolution of electronic circuits
     Evolving robot morphologies
10. Evolutionary Robotics in the real world
     A new philosophy for robotic applications
     Evolutionary Robotics for hard problems
     The case of care-giving robots

The tutorial will describe several experiments with physical evolutionary robots.
The presentation will be based on overhead projections, digital animation from computer, and several video tapes.
 

About the Lecturers

The outline of this tutorial is based on a book on Evolutionary Robotics written by Stefano Nolfi and Dario Floreano for MIT Press (to appear in autumn 2000), including more recent results not included in that book, and on several years of experience of Applied AI Systems Inc. (Takashi Gomi, President) in transferring such technology to real-world applications worldwide. The authors have already given joint tutorials on this subject at conferences such as International Conference on Artificial Neural Networks (ICANN) and European Conference on Artificial Life (ECAL).

Dario Floreano was born in 1964 (BSc. 1988; MSc. 1992; PhD. 1995). In 1993 he carried out at the Swiss Federal Institute of Technology Lausanne (EPFL) the first world experiments demonstrating artificial evolution of physical robots without human intervention. He then worked as a research fellow at the Center for Cognitive and Computational Neuroscience of the University of Stirling (UK), and later as a research scientist at the Laboratory of Cognitive Technology of the Scientific Park in Trieste (I). From 1996 to 2000 he worked as Senior Researcher at the Swiss Federal Institute of Technology in Lausanne where he was appointed Associate Professor in Adaptive and Evolutionary Robotics in 2000. In 1998 he has spent 4 months as invited researcher at Sony Computer Science Laboratories, Inc. in Tokyo. Floreano has published almost 50 technical papers and 2 books, and has presented his research at more than 70 conferences worldwide. He joined the scientific program of more than 20 international conferences, organized the 4th European Conference on Artificial Life, and is co-organizer of the next International Conference on Simulation of Adaptive Behavior (SAB2000). He is member of various international societies, of editorial board of several international journals, and served as consulting scientist for the definition of new R&D projects at the European Commission.

Stefano Nolfi was born in 1962. From 1985 to 1987 he worked as a research fellow with Domenico Parisi in the Institute of Psychology of the Italian National Research Council in Rome. He spent 6 months in 1987 as visiting researcher at the Center for Research in Language, University of California San Diego, USA. In 1988 he was hired as researcher at the Institute of Psychology in Rome where he currently he is coordinator of the Department of Neural System and Artificial Life of the Institute. Nolfi carried out some of the first computational experiments showing the interactions between learning and evolution for evolutionary robotics. Nolfi has published more than 50 papers, has presented his research at more than 50 conferences, workshops, and tutorials, and has delivered series of lectures on topics ranging from learning and development to autonomous robotics. Currently, he is leader of two projects on Evolutionary Robotics supported by the National Council of Research, Italy. He serves as expert scientist for the Research Division of the European Commission. Nolfi and Floreano have worked together on evolutionary systems at the National Research Council in 1989 and 1990, and later at the Swiss Federal Institute of Technology in 1998.

Takashi Gomi was born in Tokyo in 1940. He obtained his M. Eng in 1964 from Waseda University in Electrical and Electronic Engineering and his D. Eng in 1997 from Hokkaido University in Complex System studies.  From 1971-73 he worked as a Graduate Research and Teaching Assistant at the University of Alberta in Computing Science.  He became a member of the Scientific Staff at Bell Northern Research in 1973. After a brief experience at Atomic Energy of Canada, Mr. Gomi established Applied AI Systems, Inc. (AAI) in 1983 as a company dedicated to research and development of intelligent system technology, the oldest speciality AI company in Canada and known widely in Japan, Europe, and USA. In his company he conducts application research, intelligent system development including intelligent robots, marketing of a wide range of AI and ALife products, and training of AI research engineers.  Despite its small size (18 people), AAI is recognized as the top level intelligent system R&D organization in Canada. Since 1992, a series of R&D projects aimed at the transfer of behavior-based AI or New AI technology to government, business and industry has begun. In October 1995, Mr. Gomi became President of a newly established AI/ALife company in Japan, called AAI Japan.  Mr.Gomi is a member, IEEE Service Robot Technical Committee.

(Back to the list)

Both sensing and actuation play key roles supporting robot technology since robot motion strongly depends on their capability. Therefore, innovative sensors, sensing strategies, new actuator and actuation mechanism significantly enhance the design, development and performance of future robotic systems. Without breaking through those key technologies, it is hard for us to see highly advanced robotics systems in 21st century.  Knowing of importance of such fundamental areas, this tutorial introduces various innovative works concerning sensing and actuation.

Program:

9:00-9:10 Opening Remark

9:10-10:00 Active Wisker Sensor (Simple Principle but Numerous Application)
Makoto Kaneko (Hiroshima University)
Human can obtain various tactile information through an active motion. An interesting observation is that our finger tip can detect surface irregularities with just a few micrometer through active motions, while we can never do it without active motion. As an example of tactile based active sensing, we introduce an active whisker sensor composed of a flexible beam anchored at base, a moment sensor for getting an external information, and an actuator to move them. The application of this sensor such as burr detection after drilling, detection of surface irregularity, and measurement of the effective length of screw will be also introduced with video tape demonstration.

10:00-10:50 Telemetric Sensitive Skin
Hiroyuki Shinoda (Tokyo University of Agriculture and Technology)
The last two decades were the age of robotics with underdeveloped tactual sense. We are proceeding to the next stage fully utilizing skin sensation. In this tutorial, a new technology to realize a soft sensitive skin is introduced. An elastic body is endowed with sensation by being implanted with small tactile sensing elements which receive electrical power and transmit signal without wires. Such a skin is formed into an arbitrary shape easily, and it is elastic and tough because each sensing element needs no fragile wires. Design of the wireless sensing chip, signal processing, tactile information extracted from the signal, and fabrication and wearing of the skin will be introduced.

10:50-11:10 Break

11:10-12:00 Smart Anthropomorphic Contact Surface Technology (SACST) as Human Augmentation
Imin Kao (SUNY Stony Brook)
Research in robotics and MEMS transducers (actuators and sensors) have enabled the development of intelligent control of scalable soft surfaces.  A framework of smart anthropomorphic contact surface technology (SACST) based upon robotics theories and MEMS transducers has been developed to integrate the sensory information for feedback and control to impart intelligence.  This ongoing research has potential to revolutionize the way we design and use seats, beds, wheelchair, shoes, and artificial skin for robot arms in the future. The framework will be able to free up users from lower level tasks in order that they can concentrate on higher level tasks.  For example, if such surface is implemented in wheelchair design, the user of the chair will be able to focus on high level maneuvering without worrying about the details of empowering intelligence to the seating system.

12:00-13:00 Lunch Break

13:00-13:50 How Can High Speed Vision Change Robotics World?
Masatoshi Ishikawa (University of Tokyo)
Conventional image processing with video rate (30Hz) is too slow to directly control servo motors. High speed vision chip with 1ms processing rate has been developed and applied to high speed hand and arm. Based on the system, not only robotics world but mechanical system in general will be changed by using the high speed vision because dynamics of image processing will be matched with that of mechanics. In this tutorial, merits of high speed vision, processing architecture of vision chip, implementation using 0.35　micrometer CMOS, image processing algorithm for high speed vision, and application systems including target tracking, micro visual feedback, virtual reality, and grasping system will be explained.

13:50-14:40 Towards a Leaping/Hopping Paradigm for Planetary Exploration
Joel Burdick (Calfornia Institute of Technology)
Leaping or hopping movements are an attractive mobility alternative for small (~1 Kg size) robots that must explore low to medium gravity  planetary surfaces.  Hopping is energetically efficient in lower gravity environments, and such motions also allow small vehicles to overcome obstacles many times their body size.  This talk describes our efforts to develop small hopping planetary explorers.  After reviewing some of our previous developments, we focus on the latest two generations of our hopping vehicles.  Our second generation system uses a single motor to achieve steering, hopping, and self-righting.  Our latest generation system, under development, combines hopping for coarse long range mobility with wheels for short range precision target acquisition.  We show that a small number of actuators can control the vehicle's wheeled mobility, and hopping control systems.  The computing and communications systems are reviewed, as well as preliminary experimental results that have been obtained with our prototypes.
 
 

14:40-15:00 Break

15:00-15:50 Single Actuator Manipulation
Kevin Lynch (Northwestern University)
A single actuator can perform a variety of useful tasks involving several degrees-of-freedom.  For example, we can construct a conveyor-based parts feeder, a juggling robot, or a controllable hovercraft with a single actuator.  In this talk I will outline the theory behind these minimal actuator systems, including analysis, motion planning, and control.  The motivation is to develop robot systems which trade mechanical complexity for complexity in motion planning and control.

15:50- Closing Remark

(Back to the list)

Organizers:
Wen J. LI 1,*, Richard YEH2, Kris S. J. PISTER2, and Isao SHIMOYAMA3
1Dept. of Mechanical and Automation Engineering, The Chinese University of Hong Kong, Hong Kong
2Dept. of Electrical Engineering and Computer Science, UC Berkeley, USA
3Dept. of Mechano-Informatics, Graduate School of Engineering,, University of Tokyo, JAPAN

* Corresponding Organizer:
    The Chinese University of Hong Kong
    Dept. of Mechanical and Automation Engineering
    MMW 425, Shatin, N. T., Hong Kong
     Phone #:   +852-2609-8475      Fax #: +852-2603-6002
     E-mail: wen@mae.cuhk.edu.hk

Spirit:

More than 10 years have past since the creation of the first operable micro motor, and by now Micromaching Technology has permeated virtually all aspects of science and engineering disciplines.  However, as micro-fabrication techniques improve, micro systems are becoming much more complex and small, and an increasing amount of materials and fabrication technologies are now available to Micromachinists.  The purpose of this workshop is to introduce state-of-the-art techniques of Micromachining Technology which is currently being used to develop advanced integrated micro sensing systems that promises to impact significantly the research and development of robotic and automation activities in the coming decade.    It is our intention to bring the participants up-to-date knowledge on the materials and fabrication techniques available to integrated mechanical, thermal, chemical, optical, and electrical components and produce small and smart sensing systems, which can specifically find applications in Robotics and Automation systems.  We will do this by 1) give an overview of Micromachining/MEMS Technology and present novel materials for integrated micro systems applications; 2) present specific examples on the developmental procedures of complex micro systems, with discussions on scaling-effects, hybrid integration, assembly of small parts, and functional testing; 3) present foundry-fabrication techniques as a quick method for researchers without MEMS fabrication facilities to design and test specific sensing systems.  We will also give lectures on research and developmental projects which are at the frontier of Micromachining Technology: the development of small, intelligent, sensing systems, that merge mechanical, thermal, optical, and electrical components, and may be networked, transmit signals wirelessly, and have their own power or power-generating supply.  The advent of these systems will certainly change the robotics and automation research activities in the coming decade.

Schedule:

8:30 - 8:40    Opening Speech
Wen J. LI (The Chinese University of Hong Kong, HONG KONG)

8:40 - 9:10
Overview of MEMS and Micromachining Technology
Wen J. LI (The Chinese University of Hong Kong, HONG KONG)
A general overview of Micromachining Technology will be presented in this session.  Fundamental techniques such as surface-micromachining, bulk-micromachining, direct-energy-micromachining, and LIGA will be discussed.  Some techniques for Nanomachining will also be presented.  Emphasis is on the state-of-the-art developments for these micro/nano fabrication techniques.  Conventional and new materials for integrated mechanical, thermal, optical, chemical, and electrical micro systems will also be presented.  This session is designed to give participants an idea on the possibilities and limitations of Micromachining Technology.

9:10 - 9:40
Integrated Micro-fabricated Systems
Isao SHIMOYAMA (University of Tokyo, JAPAN)
This session will present examples of current research and development activities in complex integrated micro systems that may encompass mechanical, optical, thermal, chemical, and electrical components.  Specifically, Micromachining Technology will permit very small mechanisms to be closely integrated with control circuits, producing a complex new intelligent robot composed of microscopic actuators, sensors, batteries, circuits, and other components. Critical issues such as scaling-effects, hybrid integration, assembly of small parts, and functional testing will be discussed.

9:40 - 10:10
Integration of Foundry and Custom-fabricated Microrobotic Components
Richard YEH and Kris S. J.  PISTER (University of California, Berkeley, USA)
Using foundry fabrication allows engineers to spend more time on design and testing instead of processing.  However, in a microrobotic system, some components still require custom processing.  To take advantage of both foundry- and custom-fabrication processes, gold bump flip-chip bonding can be used.  This technique allows engineers to integrate devices and mechanisms fabricated from different processes.  A robotic system created from integrating components fabricated from MCNC and BSAC will be presented.

10:10 - 10:20  Break

10:20 - 10:50
A Small-Sized Panoramic Scanning Visual Sensor Inspired by the Fly's Compound Eye
Isao SHIMOYAMA (University of Tokyo, JAPAN)
This session will present our investigation of the feasibility to design a small artificial compound eye that comprises micro-optical, micro-mechanical components and additional electronics for visual processing. The hybrid integration of this "mini-sensor" addresses important issues such as mechanic scale-effect, assembly of small parts and functional testing.  It was recently discovered that the fly's retina undergoes repetitive scanning vergence movements. The use of retinal scanning for the perception of visual motion is highlighted in the design of a small-sized horizontal array with a scanning retina of 2x30 photoreceptor elements (Reye = 30mm). The scanning motion is achieved by actuating the photodiode arrays with a piezoelectric bimorph cantilever beam coupled to a plate spring: the system can be brought to a resonant mode at 100Hz (scanning amplitude > 80um) under low voltage (DV0 V). The mini-compound eye, equipped with analogue Motion Detecting circuits, is able to guide the steps of a small mobile platform and perform multi-task visual processing for obstacle avoidance, tracking and pursuit of visual target. The visual sensor is a first step in the evaluation of the design and development of a micro-visual system whose size will eventually approach the size of the fly's compound eye.

10:50 - 11:20
Low-Power Actuators for Microrobots
Richard YEH and Kris S. J.  PISTER (University of California, Berkeley, USA)
Power and size are important considerations when it comes to making actuators for autonomous microrobots.  On the microscale, electrostatic gap-closing actuators are an attractive option due to their low power consumption, high power density, and efficiency.  Although, these actuators have a force to displacement trade-off, they can be made in an inchworm-drive configuration to decouple force from total displacement. A low-power electrostatic inchworm motor with large force and total displacement will be presented.

11:20 - 11:50
Wireless Sensors with Integrated Vibration-induced Power Generator
Wen J. LI (The Chinese University of Hong Kong, HONG KONG)
The advent of MEMS/micromachining technologies is enabling the research and development of distributed sensors and control systems.  These systems will permit fundamental advances in manufacturing automation and condition-based maintenance. As an example, distributed sensing systems for manufacturing include micro sensors installed on rotating tool bits, machine tools, and workpieces to measure feed-force, vibration, and stamping.  Ideally, each sensor module of a distributed sensing system should be networked, transmit signals wirelessly, and contain its own power source, so that the entire system can be mobile and reconfigurable to prevent maintenance downtime which can be excessively expensive.  Our goal is to demonstrate an integrated sensing module capable of wireless transmission and on-chip power generation.  We have, thus far, developed a foundry surface-micromachined high-speed rotation sensor which is able to wirelessly transmit data from 100 to 6000rpm using a commercial transmitter, and a laser-micromachined vibration-induced power generator with 2.0V DC output which has enough power to drive an IR transmitter to  send 140ms pulse trains with ~60seconds power generation time.  The design, analysis, and results of these devices will be present at this session.

11:50 - 12:00  Closing Remarks
Wen J. LI (The Chinese University of Hong Kong, HONG KONG)
 

(Back to the list)