2001 Events:
November 15, 2001: "Hierarchical Approach for Design of Multi-Vehicle Multi-Modal Embedded Control Systems" by Dr. T. John Koo, University of California at Berkeley
Abstract: Embedded systems composed of hardware and software components are designed to interact with a physical environment in real-time in order to fulfill control objectives and system specifications. In this talk, we address the complex design challenges in embedded control systems by focusing on predictive and systematic hierarchical design methodologies which promote system verification and validation. First, we advocate a mix of top-down, hierarchical design and bottom-up, component-based design for complex control systems. Second, it is our point of view that at the level closest to the environment under control, the embedded software needs to be time-triggered for guaranteed safety; at the higher levels, we advocate asynchronous hybrid controller design. We briefly illustrate our approach through an embedded software design for the control of a group of autonomous vehicles.
bio: T. John Koo received the B.Eng. degree in 1992 in Electronic Engineering and the M.Phil. in 1994 in Information Engineering both from the Chinese University of Hong Kong, and the Ph.D. degree in 2000 in Electrical Engineering and Computer Sciences from the University of California at Berkeley.
From 1994 to 1995, he was a graduate student researcher in Signal and Image Processing Institute at the University of Southern California. He was a consultant of SRI International in 1998. From 1996-2000, he was the project leader and founder of the Berkeley Aerial Robot Project. In fall 2000, he was a postdoctoral researcher in GRASP Laboratory at the University of Pennsylvania. Currently, he works as a specialist in the Electronics Research Laboratory at the University of California at Berkeley. His research interests include nonlinear control, hybrid systems, embedded software, inertial navigation systems with applications to unmanned aerial vehicles.
He received the Distinguished M.Phil. Thesis Award of the Faculty of Engineering, the Chinese University of Hong Kong, in 1994. He is a member of IEEE and SIAM.
October 18, 2001: "Advancing the Art of Minimally Invasive Surgery through Robotics: An Overview of the da Vinci Surgical System" by Mr. Mike Tierney, Intuitive Surgical, Inc.
Abstract: In breakthrough operations performed in the early 1980s, minimally invasive surgical (MIS) techniques showed tremendous promise in reducing patients? pain and recovery times when compared to the more traumatic open surgery techniques for similar procedures. MIS techniques are based upon allowing the surgeon to operate through tiny ports (less than 1 cm puncture wounds) using specially designed instruments and endoscopes. However, largely due to technological limitations, MIS concepts were only adopted to those procedures that did not require a high degree of delicate, precise movement in the operative field. Unfortunately, this represented only a very small percentage of those procedures that could have benefited from MIS, thereby causing early advancement of MIS techniques to reach a premature plateau.
In the late 1980's, research at both SRI International and IBM Corporation pioneered proof of concept work that demonstrated robotics could be used to overcome the fundamental limitations of MIS techniques. Founded in 1995, Intuitive Surgical, Inc. acquired exclusive licenses from both SRI and IBM to expand and enhance their early work. Additionally, Intuitive teamed with researchers at MIT to develop a miniature mechanism that added three degrees of freedom in the proximity of the instrument tip, which was to operate within the patient's body. This invention led to a family of remotely operated MIS surgical instruments that would achieve the unprecedented faithful replicative movements of the surgeon's wrist and hand while s/he operated at a distant console. The initial fruit of these collaborative and intense internal design efforts within Intuitive is a surgical teleoperator system, known as the da Vinci Surgical System, that is replete with a suite of interchangeable surgical instruments. Designed to control several multi degree-of-freedom manipulator arms, this system is now becoming one of the surgeon's most valuable tools in the performance of complex surgical procedures. In the United States, the da Vinci System so far has been FDA cleared for laparoscopic surgical procedures (July, 2000), thoracic surgical procedures (March, 2001), and radical prostatectomys (June, 2001).
This presentation will consist of an overview of the da Vinci Surgical System from an engineering perspective. It will also discuss the engineering problems that had to be solved for such a system to overcome the limitations of existing MIS techniques, to gain initial FDA clearance, and to be readily accepted by surgeons.
bio: Mike Tierney is a Principal Engineer at Intuitive Surgical in Mountain View, California, where he joined in 1996 and served as the daVinci System Hardware Architect. Prior to Intuitive, from 1991 to 1996, he was a Senior Member of the Technical Staff at Acuson, where he contributed major imaging and video processing designs to the Aspen Ultrasound System. From 1985 to 1991 he was a Staff Engineer at Siemens Ultrasound where he contributed user interface, physiological controls, and beam former control designs to the Sonoline Series of Ultrasound Systems.
He received his BEEE degree from Vanderbilt University in 1975, where he graduated Summa Cum Laude. He received his MSEE degree from Santa Clara University in 1985. He is an active member of the IEEE in the Computer, Control Systems, and the Engineering in Medicine & Biology Societies, and is a member of the Tau Beta Pi and Etta Kappa Nu Honor Societies.
June 21, 2001: "An Embedded Systems Approach to Closed-Loop Control Systems" by Ms. Patricia O'Neil, Microchip Technology Inc.
Abstract: Current developments are bringing Digital Signal Processing (DSP) techniques into the microcontroller arena. The purpose of this talk is to introduce microcontroller basic concepts and development tool environments for control systems applications. We will then build off these basic constructs to look at advanced digital techniques such as moving average filtering, sampling as mixing and Proportional-Integral-Derivative (PID) control loops as they apply to embedded control systems. Examples of these basic DSP functions will be given using the PIC18CXXX family of microcontrollers.
bio: Patricia O'Neil is a Field Applications Engineer at Microchip Technology, Inc., in San Jose, CA. She received her B.S. in electrical engineering from San Jose State University in San Jose, California, with a focus in microcontrollers. Patty is the Santa Clara Valley IEEE Graduates of the Last Decade (GOLD) Chairperson.
May 17, 2001: "Modular Reconfigurable Robotics" by Dr. Mark Yim, Xerox Palo Alto Research Center
Abstract: Modular Reconfigurable Robots are robots built from many copies of a few simple module types (similar to Lego bricks, or cells in mammals). A modular robot that can reconfigure itself -- change its shape by moving its modules around -- can remove failing modules and reconfigure to meet the demands of changing tasks and environments. As the number of modules increases from tens to hundreds to thousands, these systems show promise of great versatility, robustness and low cost. However, to make this realizable there are many computation, control and manufacturing issues that must be addressed.
We will show progress on several modular reconfigurable robot systems developed at the Palo Alto Research Center with videos of a variety of locomotion and manipulation tasks and present some of the issues in applying them to urban search and rescue and shape configuration. These tasks are rich in interesting problems including: distributed computation and control, modular design, reconfiguration planning, motion planning, computational geometry, and many others. https://www.parc.xerox.com/modrobots
bio: Mark Yim received his PhD in mechanical engineering from Stanford University in 1994. He then worked at a start-up making force feedback devices for virtual reality applications for several years before moving to the Xerox Palo Alto Research Center (PARC) where he is a senior member of the research staff.
Currently, he leads the modular robotics team at PARC. The study of how to build and control this type of machine is part of PARC's research on "Smart Matter", which is at the intersection of the fields of Computer Science, small electro-mechanical systems and distributed control among others.
He was chosen as one of the TR100, the top 100 young innovators by MIT's Technology Review Magazine. He has authored over 20 patents and published in the areas of mobile robot planning, distributed robotics, optimal control, MEMS, and haptic devices. His work has been featured in a variety of popular press (New York Times, MSNBC, ABC and CBS News, USA Today, and various other local and international news media and strangely enough, a woman's fashion magazine.)
April 17, 2001: "Latency Considerations in High Bandwidth Control Loop Design: The Space Interferometry Mission Path Length Control System" by Dr. David Schaechter, Lockheed Martin Advanced Technology Center, Palo Alto, California
Abstract: The Hubble Telescope, the Next Generation Space Telescope (NGST), the Space InfraRed Telescope Facility (SIRTF), the Space Interferometry Mission (SIM), and many others are all part of NASA's Origins Program. Currently JPL has teamed with two Industry Partners to design, develop, integrate and test a "telescope" that looks at the universe in a new way, through the use of interferometry, this mission being named SIM. The goals of SIM are to provide micro arcsecond astrometric measurements of the universe, which among other things, is sufficiently accurate to infer the presence of earth-size worlds orbiting nearby stars, through indirect observation of the planets' reflex motions on the central stellar bodies.
In order to obtain this level of astrometric resolution, SIM's sensing and control performance numbers reside in the world of nanometers and picometers. This presentation will give a top-level overview of the SIM Mission and the instrument operation. Additional time will focus on one particularly interesting aspect of the control problems associated with SIM, that of the effects of time delays in closing high bandwidth control loops.
bio: Dr. David Schaechter is currently a Consulting Scientist and Senior Member of the Research Laboratory at the Lockheed Martin Advanced Technology Center where he is group leader for the Control System Technology Group that consists of 15 world-class dynamics and controls experts. He was formerly a member of the technical staff at JPL, and an Acting Assistant Professor at Stanford University.
Dr. Schaechter's present assignment is the Lockheed Martin lead for the Space Interferometry Mission (SIM) Real Time Controls effort. He has worked other programs such as the Space-Based InfraRed System (SBIRS) Dem Val and EMD payload pointing control assembly brassboard efforts, and has been involved with numerous other precision pointing and control programs, including LODE, TALON GOLD, ABCS, AWS, BSTS, FEWS, NASA's New Millennium and Autonomous Rendezvous and Capture Programs, several classified and IRAD programs, and LM launch vehicle guidance.
Dr. Schaechter is the originator and developer of AUTOLEV multibody dynamics software and the author of 28 publications. He served 9 years on AIAA Astrodynamics, and Guidance and Control Technical Committees. He is the recipient of 2 NASA Awards, 3 LM Awards, AIAA Engineer of the year Section Award, and AAS Member of the Month. David is a Fellow of AAS and an Associate Fellow of AIAA and has served as Associate Editor of the AIAA Journal of Guidance and Control, and as Editor of the Journal of Astronautical Sciences.
March 15, 2001: "An LMI Approach to the Control of a Compact Disc Player" by Marco Dettori, SC Solutions, Inc.
Abstract: In a Compact Disc Player a highly accurate control of the tracking servomechanism is required. In the absence of physical contact between the readout device and the disc, a position controller has to guarantee that the laser spot follows the track where the information is stored. The requirement on the position error between laser spot and track is a hard bound on its time-domain amplitude. In current audio applications this quantity cannot exceed 0.1 microns. Emerging high-performance applications, like CD-ROM or DVD-ROM, require a higher data density on the disc and a faster data readout, resulting in a more severe bound for the error amplitude. On the other hand, manufacturing tolerances in mass-production generate variations from player to player, leading to the necessity of designing robust controllers. The purpose of this talk is to show to what extent the performance of the CD player mechanism can be enhanced, in the presence of model uncertainty, using recently developed design techniques based on Linear Matrix Inequalities (LMIs). In particular, we will focus on the LMI approach to the design of gain-scheduling controllers and we will present some experimental results of digital controller implementation.
bio: Marco Dettori is a Senior Research Engineer at SC Solutions, Inc. in Santa Clara, California. He received his M.Sc. in electrical engineering from the University "La Sapienza" of Rome, Italy, in 1994 with a thesis on nonlinear H-infinity control. From January 1996 to April 2000 he has been a Ph.D. student at the Mechanical Engineering Systems and Control Group of Delft University of Technology in the Netherlands. His research project was focused on the development of robust control techniques based on Linear Matrix Inequalities and their application to Optical Disc Drive systems and was performed in collaboration with Philips Research Laboratories, Eindhoven, the Netherlands. In 1997, he obtained a postgraduate specialization school certificate from the Dutch Institute of Systems and Control (DISC). Marco Dettori will have his Ph.D. defense in Delft on April 10.
February 15, 2001: "Control and Estimation Issues in Integrated Driver Assistance Systems" by Prof. J. Christian Gerdes, Mechanical Engineering - Design Division, Stanford University
Abstract: The actuation and sensing capabilities of the average passenger car continue to expand, opening up new possibilities for diagnosis and control. While one logical interpolation of this trend is complete automation, a number of issues - not the least of which being that driving can be great fun - suggest that the ideal level of human involvement is something other than zero. To formulate driver assistance systems that rely heavily on control while leaving the driver in the loop, better vehicle state and parameter estimation and a coherent method for coordinating vehicle control subsystems are required. This talk discusses some of the current work in these areas in the Dynamic Design Lab at Stanford including the use of GPS velocity information for determining tire parameters and an integrated approach to assistance using artificial potentials.
bio: J. Christian Gerdes is an Assistant Professor in the Design Division of the Mechanical Engineering Department at Stanford University. He received a B.S. in Mechanical Engineering and Applied Mechanics, a B.S. in Economics and an M.S. in Mechanical Engineering from the University of Pennsylvania and a Ph.D. in Mechanical Engineering from the University of California at Berkeley. He has been involved with vehicle dynamics and control for the past nine years, beginning with research in coordinated throttle and brake control in the California PATH program while pursuing the Ph.D. After receiving the Ph.D., he joined Daimler-Benz and founded the Vehicle Dynamics research group at the Vehicle Systems Technology Center in Portland, OR. While at Daimler, he developed vehicle dynamics simulation methods and participated in vehicle design projects with Freightliner Corporation, the largest manufacturer of heavy trucks in North America. The results of this work can be seen in several products including the Freightliner Panther FL Rapid Intervention Vehicle and the Tuf-Trac Vocational Suspension. Since 1998 he has led the Dynamic Design Lab at Stanford University. He teaches courses in the areas of vehicle dynamics, machine design and system identification.
January 18, 2001: "Control of Many-Element Systems: An MEMS airjet based system" by Dr. Warren B. Jackson, Xerox Palo Alto Research Center (PARC)
Abstract: Control techniques are typically applied to systems consisting of a few sensors, actuators and controllers. In this talk, closed loop control (loop time-1ms) of an airjet system consisting of 1000 airjet actuators and 30000 sensors fabricated using printed circuit board based MEMS is presented. In such many-element systems, such issues as actuation allocation, fusion of sensor data, and system identification emerge as interesting areas of research. Solutions for allocating actuation among large number of actuators using hierarchical constrained optimization and fusing the output of many sensors into a small number of final measurements under tight real time constraints will be presented. Most interesting, such a hyper redundant system is capable of self-system identification where the aggregate of properties of many elements can be used to measure detailed properties of individual elements. The refined measurements of individual properties can then be used to improve estimates of the aggregate. This novel capability as well as the standard performance characteristics of the system control will be presented. This talk presents joint work with M. Fromherz, J. Reich, B. Preas, D. Biegelsen, L. Schwartz, D. Goldberg, and A. Berlin.
bio: Warren Jackson is currently Principal Scientist at Xerox Parc. He received his undergraduate at Stanford and PhD at Berkeley in Physics. He has published over 140 papers in the fields of control of many element systems, image quality metrics, smart pixels, large area electronics, and the physics of disordered materials. He has been granted over 35 patents and is a fellow in the American Physical Society.